A P aradigm f o r G e n era l S ce n e A n a ly sis.
1 s t O c t o b e r , 1985
D a v id A le x a n d e r F o r s y t h .
DEDICATION
DEDICATION.
I d e c la re t h a t t h i s
d is s e r ta tio n
is
t h e r e s u l t o f cay own u n a id e d w o rk ,
w h ic h h a s n o t b e f o r e b e e n s u b m i tte d t o a n y u n i v e r s i t y o r i n s t i t u t i o n f o r
a n y d e g r e e o r d ip lo m a , s a v e t h a t w o rk c l e a r l y d e s c r i b e d
h a s been in s e r te d
f o r th e p u rp o se o f c o n tin u ity .
and c o r r e c te d , and i s f r e e o f e r r o r s .
SIGNED:
Declaration.
a s s u c h , w h ic h
I t h as been p ro o frea d
A
d is s e rta tio n
s u b m i tte d
to
th e
U n iv e rs ity
of
th e
W itw a te r s r a a d
f u l f i l m e n t o f t h e r e q u i r e m e n t s f o r t h e d e g r e e o f M a s te r o f
E l e c t r i c a l E n g in e e r in g .
S c ie n c e
in
in
A f a r m e r , w a n t in g a n a u to m a t ic c h ic k e n p l u c k i n g m a c h in e , s p o k e t o a n e n ­
g i n e e r , who sw o re t h a t i t c o u l d n 't b e b u i l t .
w e n t t o a t h e o r e t i c i a n , who s a i d " i t ' s
a r b i t r a r y d ia m e t e r d ,
...”
(M odern F o l k t a l e , o f unknow n o r i g i n . )
The d is a p p o in te d fa rm e r th e n
e a s y ! A ssume a s p h e r i c a l c h i c k e n ,
ACKNOWLEDGEMENTS.
T h e f o l l o w i n g p e o p le a n d c o m p a n ie s h a v e h e lp e d m e,
a n d I th a n k th em f o r
t h e i r tim e , p a tie n c e , and s u g g e s tio n s .
A d v an c ed S e m ic o n d u c to r D e v ic e s l e n t me a n im a g e a c q u i s i t i o n
and d is p l a y
s y s te m , w i t h o u t w h ic h I c o u ld h a v e d o n e v e r y l i t t l e .
T h e C o u n c il f o r M in e r a l T e c h n o lo g y p r o v id e d g e n e r o u s f i n a n c i a l s u p p o r t .
H e w le tt- P a c k a r d
(S .A .)
l e n t me a m in ic o m p u te r s y s te m , o n w h ic h a l l t h e
w o rk d e s c r i b e d h e r e , a n d a l o t m o re , w as d o n e .
M i t s u i C o r p o r a t i o n p r o v i d e d an e x c e l l e n t s o f t w a r e p a c k a g e , SPIDER,
im a g e p r o c e s s i n g
fo r
a p p l i c a t i o n s , w h ic h h e l p e d me o v e r a n im m ense l e a r n i n g
T h e s t a f f a n d s t u d e n t s a t U w tech p r o v id e d a v e r y p l e a s a n t w o rk in g e n v i ­
r o n m e n t, w h ic h m ade my w o rk e a s i e r .
G u n th e r Sommer a n d P r o f e s s o r B ru c e
S t e w a r t o f t h e C o u n c il f o r M in e r a l T e c h n o lo g y g a v e me t h e i r s u p p o r t a n d
i n t e r e s t , a n d w e re v e r y t o l e r a n t o f my w i l d e r i d e a s .
D ave P e n c k l e r a n d A d ria n Solom on o f H e w le tt- P a c k a r d h e lp e d me t o u s e
th e
m in i c o m p u t e r , a n d r e s c u e d me o n a n u m b er o f o c c a s i o n s .
G e r a l d B lo c h a n d S t e v e M eyer o f U w tech p r o v i d e d s t i m u l a t i n g d i s c u s s i o n ,
a n d much s u p p o r t .
Acknowledgements.
1
ACKNOWLEDGEMENTS.
T he f o l l o w i n g p e o p le a n d c o m p a n ie s h a v e h e lp e d m e,
a n d I th a n k th e m
fo r
t h e i r tim e , p a tie n c s , and s u g g e s tio n s .
A dvan c ed S e m ic o n d u c to r D e v ic e s l e n t me a n im a g e a c q u i s i t i o n
and d is p la y
s y s te m , w i t h o u t w h ic h I c o u ld h a v e d o n e v e r y l i t t l e .
T he C o u n c il f o r M in e r a l T e c h n o lo g y p r o v id e d g e n e r o u s f i n a n c i a l s u p p o r t .
H e w le tt- P a c k a r d
(S .A .)
l e n t me a m in ic o m p u te r s y s te m , o n w h ic h a l l t h e
w o rk d e s c r i b e d h e r e , a n d a l o t m o re , w as d o n e .
M i t s u i C o r p o r a t i o n p r o v i d e d a n e x c e l l e n t s o f t w a r e p a ck a g es, SPIDER,
fo r
im a g e p r o c e s s i n g a p p l i c a t i o n s , w h ic h h e lp e d me o v e r a n im m ense l e a r n i n g
T h e s t a f f a n d s t u d e n t s a t U w tech p r o v id e d a v e r y p l e a s a n t w o rk in g e n v i ­
r o n m e n t, w h ic h made my w o rk e a s i e r .
G u n th e r Sommer a n d P r o f e s s o r B ru c e
S t e w a r t o f t h e C o u n c il f o r M in e r a l T e c h n o lo g y g a v e me t h e i r s u p p o r t a n d
i n t e r e s t , a n d w e re v e r y t o l e r a n t o £ my w i l d e r i d e a s .
D ave P e n c k l e r a n d A d r i a n Solom on o f H e w le tt- P a c k a r d h e lp e d
me t o
u se th e
m in i c o m p u t e r , a n d r e s c u e d me o n a n u m b er o f o c c a s i o n s .
G e r a l d B lo c h a n d S t e v e M eyer o f U w tech p r o v id e d s t i m u l a t i n g d i s c u s s i o n ,
a n d much s u p p o r t .
Acknowledgements.
ill
D av e S m ith a n d G u n th e r B e r g e r , who w o rk e d w i t h me i n t h e im ag e p r o c e s s i n g
g r o u p , g e n e r a t e d a s t i m u l a t i n g w o rk in g e n v iro n m e n t a n d a llo w e d me t o ma­
t u r e my i d e a s th r o u g h lo n g d i s c u s s i o n s .
a s th e re a d e r w ill s e e ,
My d e b t t o G u n th e r i s v e r y l a r g e ,
an d h e h a s b e e n a c o n s i s t e n t l y
stim u la tin g
re ­
s e a r c h p a r t n e r s i n c e we s t a r t e d w o rk in g t o g e t h e r i n e a r l y 1 9 8 4 .
M rs. S u e Rodd v e r y k i r d l y p r o o f r e a d t h i s w o rk i n a v e r y s h o r t t i m e , a n d
d Id s o v e r y w e l l .
Any e r r o r s w h ic h r e m a in i n t h e t e x t a r e my f a u l t e n -
P r o f e s s o r M ike R o d d , my s u p e r v i s o r , h a s g iv e n me tr e m e n d o u s s u p p o r t a n d
e n c o u r a g e m e n t,
a lo n e
rep ay ,
a n d I owe h im a d e b t t h a t i s
fo r h is
a d v ic e ,
d if f ic u lt to
h is p a tie n c e , h is
e x p re s s ,
le t
e n c o u r a g e m e n t, a n d t h e
s u p p o r t h e g a v e me.
My f a m i ly h a s - t o l e r a t e d me w hen t h e w o rk made me i n t o l e r a b l e ,
an d I owe
th e m g r e a t t h a n k s f o r t h i s .
F i n a l l y , T e s s v o n W i l l i c h d rew t h e p i c t u r e s , made t h o s e c h a n g e s i n t h e
t e x t t h a t I w as s u g g e s t e d , c o o k e d , k e p t h o u s e , r e m a in e d t o l e r a n t a n d g o od
n a t u r e d , a n d b e l e i v e d I was d o in g w o r t h w h i le w o rk w hen I w as d o u b tin g i t .
F o r t h i s h e l p , I owe h e r a tr e m e n d o u s am o u n t.
Acknowledgements.
a bstra ct
A new p a ra d ig m
is
d e s c rib e d
p a ra d ig m r e q u i r e s f i r s t l y
fo r
k n o w le d g e - b a s e d
sc en e
a n a ly s is .
T he'
t h a t im a g e s b e u n d e r s t o o d e s s e n t i a l l y i n te rm s
o f tw o - d im e n s io n a l i n f o r m a t i o n a b o u t how o b j e c t s
Jn p ic t u r e s
s e c o n d ly t h a t f e a t u r e s b e c o n s i d e r e d a s c o l l e c t i o n s o f r e g i o n s .
f o r th e a n a ly s i s o f g e n e r a l sc en e s i s
lo o k ,
d e s c r ib e d and th e r e s u l t s
a t t e m p t t o p r o t o t y p e t h i s s y s te m a r e g i v e n .
and
A s y s te m
of
en
A lth o u g h t h e p r o t o t y p e i s a
s i m p l i f i e d v e r s i o n o f t h e s y s te m d e s c r i b e d , i t a p p e a r s t o b e c a p a b l e o f
g r e a t p o w er.
T h e p a ra d ig m i s f u r t h e r s u p p o r te d b y p r o m is in g r e s u l t s o b ­
t a i n e d I n w o rk o n a n in s t r u m e n t t o a n a l y s e t h e s i z e d i s t r i b u t i o n o f p a r ­
t i c l e s on a c o n v e y o r b e l t .
A b s tr a c t
TABLE OF CONTENTS
IN TRODUCTION...................................................................... : ...........................................
BACKGROUND
......................
1
3
T h e o rig in s o f t h i s w o rk ...............................................................................................
4
K n o w le d g e -b a se d s y s te m s ..............................................................................................
6
S c e n e -a n a ly s is p a r a d ig m s .............................................................................................. 10
Im age r e p r e s e n ta t i o n ...............................................................................................
K now ledge r e p r e s e n ta tio n
13
.......................................................................................... 16
K now ledge m a n ip u la tio n .............................................................................................
.
18
R e a so n in g a b o u t c o n tr o l
.
20
, .
Im plem entation i s s u e s ............................................................................................................22
C o n c lu s io n ....................................................................................................
22
NEIGHBOUR CO D ES............................................................................................................... 23
P r o p e r tie s o f t h e n e ig h b o * C ode.................................................................................. 24
F orm ing t h e n e ig h b o u r c o d e .............................................................................
25
C o n c lu sio n ...........................................................................................
31
32
THE PARADIGM AND ITS U SE..........................................................
G u id in g c o n c e p t? >’ t h e p a ra d ig m its e lf ...........................
- ............................... 32
A sy ste m d e s ig n e d to a n a ly s e sc e n e s
- -
P o te n tia l p ro b le m s
C o n clu sio n
.
. . . . '.........................................................................................
THE DESIGN OF THE PROTOTYPE SYSTEM.
T ab le o f C o n ten ts
35
.......................................................................
40
41
T h e block d e s ig n of t h e p r o to ty p e sy stem .
T h e d e sig n of th e d a ta s t r u c t u r e s .
. .
T h e s y n ta x a n d s t r u c t u r e o f th e know ledg
E x am p les o f r u l e u s e ................................
T h e alg o rith m s u s e d w ith in t h e b lo c k s.
How t h e p r o t o t y p e e v a l u a t e s a K S.
How t h e p r o t o t y p e f a c t o r s a g o a l .
How t h e s y s te m g e n e r a t e s a n d t e s t :
F in d in g th e g o a l.
C o n c lu sio n .
. . .
THE IMPLEMENTATION OF THE PROTOTYPE.
E r r o r h a n d lin g ...................................................
C o d in g te c h n iq u e s u s e d f o r b u ild in g
RESU LTS."
...................................................................
R e s u lts o b ta in e d w ith in t h e p a ra d ig m .
. .
R e s u lts o b ta in e d b y p ro to ty p in g th e sy ste m .
CONCLUSION
REFERENCES.
,
R E A D IN G S ............................................................................................................................................ 91
A P P E N D ,C E S
.................................................................................................................................92
A P P E N D IX A . T H E S U B R O U T IN E S IN V O L V E D
P ro g ra m RINT
.................................................. 9 3
............................................................................................................................ 93
P ro g ra m IN IT S
.................................................................................................................
P ro g ra m START
...........................................................
94
94
P ro g ra m FGOAL
..................................................................................................................
95
P ro g ra m GTMER
....................... •...............................................................................................96
S u b r o u t i n e AGET
...........................................................
97
S u b r o u ti n e AHEAD
............................................................................................................
97
S u b r o u ti n e ANDSTACK
..................................................................................................
98
S u b r o u t i n e APSETUP
.......................................................................................................
98
S u b r o u t i n e ASETUP
........................................................................................................
S u b r o u ti n e ATORSTACK
......................................................
98
99
S u b r o u t i n e BACKE
............................................................................................................
S u b r o u t i n e B IS
.......................................................................................................
100
..............................................................................................
100
S u b r o u t i n e CAN’DSTACK
S u b r o u t i n e CHCTXT
S u b r o u ti n e CLEAN
...................................................................................................
........................................................................................
99
100
101
S u b r o u t i n e CLFIND
' .
101
S u b r o u t i n e CMERGE
...................................................................................................
101
....................................................................................
102
S u b r o u ti n e CONHEAD
S u b r o u ti n e COMPLIST
..................................................................................................
.
102
103
S u b r o u ti n e CONMAT
......................................................
S u b r o u ti n e CSSETUP
.......................................................................... . . . . .
103
S u b r o u ti n e CVARFCL
..................................................................................................
104
S u b r o u ti n e CVARPIX
. .
105
Table of Contents
. . . . . . . . .
. * ..........................................................
v
S u b r o u ti n e
S u b r o u ti n e QUALEVAL
RQUEUE
•' ‘ ■
S u b r o u t i n e SEGRT
.......................................................................................................
121
S u b r o u ti n e SIMGE
.......................................................................................................
121
S u b r o u ti n e SIMTE
S u b r o u t i n e STRLIST
S u b r o u t i n e TRUEB
S u b r o u ti n e VECREVA1
APPENDIX B .
...................................................................................
122
..................................................................................................
122
.......................................................................................................
123
.............................................................................................
123
THE BACKUS
NAUR FORM SYNTAX OF THE KS’S .
125
APPENDIX C . A COMPLETE LISTING OF THE CODE...............................
126
• T a b l e o f C o n t e n ts
LIST OF FIGURES.
F ig u re 1 :
T h e R i t e o f S p r in g .
F ig u re 2:
A N e ig h b o u r Code o f t h e Form
F ig u re
3:
T he
R eason
th e
D ire c tio n
12
of
3 4 ,3 4 5 .
P ix e l
T ra v e rsa l
in
F o rm in g
N e ig h b o u r C o d e s i s I m p o r t a n t .
F ig u re 4 :
& Code T h a t C o n t a c ts i n N + l P o i n t s E n c l o s e s N S e t s o f R e g io n s .
F ig u re 5 :
To I l l u s t r a t e M e rg in g R e g io n s
F ig u re 6 ;
T h e B lo c k D ia g ra m o f a S y stem D e s ig n e d U n d er t h e New P a r a d ig m .
F ig u re 7:
T h e B l o c k D ia g ra m o f t h e P r o t o t y p e S y s tem .
F ig u re S:
T o I l l u s t r a t e E v a l u a t i n g K S 's .
fig u re 9:
A P i c t u r e T a k e n a t D e e l k r a a l M in e , an d i t s S e g m e n ta t io n b y t h e
F ig u re 10:
T h e K n o w led g e B a se U sed i n t h e E x am p le.
F ig u re 11:
T h e P i c t u r e U sed i n t h e E x a m p le , a n d i t s R e p r e s e n t a t i o n .
F ig u re 12:
T h e O u tp u t o f t h e P r o t o t y p e S y stem f o r t h e E x am p le.
List of figures.
. 1
INTRODUCTION.
T h is d i s s e r t a t i o n
o ffe rs
a new p a ra d ig m
f o r sc en e a n a ly s is ,
th e
b a s ic
t e n e t s o f w h ic h a r e t h a t a s c e n e s h o u l d b e a n a ly s e d u s i n g e x p l i c i t know ­
le d g e , a n d t h a t t h e s c e n e s s h o u ld b e u n d e r s t o o d a n d r e p r e s e n t e d i n te rm s
o f re g io n s .
We w i l l d i s c u s s t h e o r i g i n s
a nd im p lic a tio n s o f th e s e t e n e t s ,
som e r e s u l t s w h ic h i n d i c a t e t h a t t h e y a llo w
c a n in d e e d a n a l y s e s c e n e s .
a s y s te m t o
a n d show
b e b u i l t w h ic h
T he a d v a n ta g e s o f t h e u s e o f r e g i o n s t o i n ­
t e r p r e t t h e s c e n e w i l l b e show n t o b e c o n s i d e r a b l e ,
le d g e t o b e a c t i v e l y d e p lo y e d .
a s th e y
a llo w know ­
We w i l l d i s c u s s a r e p r e s e n t a t i o n d e v e lo p e d
t o h a n d le t h e p i c t u r e a s a s e t o f r e g i o n s , an d show how t h e -
’io n s may
b e h a n d le d u s i n g t h i s
re
r e p r e s e n t a t i o n , w h ic h
is
.known
as
:;
n e ig h b o u r c o d e .
A s y s te m t h a t i s
c a p a b le o f g e n e r a l s c e n e a n a l y s i s i s d e s c r i b e d , a n d we
d i s c u s s t h e p r o t o t y p i n g o f a s im p le v e r s i o n o f s u c h a s y s te m ,
o f th e p ro to ty p e i s
e x h a u s tiv e ly p r e s e n te d ,
an d
its
T he d e s ig n ,
i m p l e m e n ta tio n
is
d i s c u s s e d . T he c o d e o f t h e p r o t o t y p e s y s te m i s p r e s e n t e d a s a p p e n d ic e s .
T h e a p p ro a c h p r o p o s e d c a n n o t b e show n t o b e th e b e s t o n e , n o t c a n i t y e t
b e show n t h a t i t w i l l d e f i n i t e l y w o rk , b u t we h a v e m anaged t o show t h a t
a s y s te m t h a t w o rk s c a n b e d e v e lo p e d w i t h i n t h e p a ra d ig m , a lth o u g h i t i s
n o t y e t 'a n a l y s i n g
h o w e v e r, t h a t i t
re a l scen es.
We w i l l p r o d u c e
c o n v in c in g
i s a g o o d a p p r o a c h , a n d t h a t i t c a n w o rk .
a rg u m e n ts ,
R e s u l t s b o th
o f t h e p r o t o t y p i n g o f a g e n e r a l s c e n e a n a l y s i s s y s te m a n d o f w o rk o n a n ­
INTRODUCTION.
o th e r system which is claim ed t o have been developed w ith in t h i s paradigm ,
in d i c a t e t h a t w orking system s may be b u i l t w ith in th e s e t e n e t s .
INTRODUCTION.
BACKGROUND
C o m p u te r v i s i o n p r e s e n t s a f a s c i n a t i n g p r o b le m , w i t h many p o t e n t i a l e n ­
g in e e r in g b e n e f its p r e s e n te d by even a p a r t i a l s o lu t io n .
t h e p r o b le m t h a t we s h a l l a d d r e s s i s
T he a s p e c t o f
t h a t o f s c e n e a n a l y s i s , w h ic h c a n
f a i r l y b e d e s c r i b e d a s a t t e m p t i n g t o s o l v e t h e p r o b le m :
'.'G iv en a p i c t u r e , d e s c r i b e w h a t i s h a p p e n in g i n i t . "
o r th e r a t h e r s im p le r v e rs i o n :
" G iv e n a p i c t u r e , f i n d t h e
------ i n i t . "
T h e s e c a n b e s e e n t o b e s u b t l e p r o b le m s .
as i t
We s h a l l a d d r e s s t h e l a t t e r ,
i s a s im p le r v e r s i o n o f t h e fo rm e r, a n d h a s n o t y e t b e en s o lv e d w ith
a n y d e g r e e o r p r o m is e o f g e n e r a l i t y .
Many t e c h n i q u e s h a v e b e e n p r o p o s e d t o s o l v e t h i s
m o st s o l u t i o n s b e i n g t h e f i n d i n g o f a p p r o p r i a t e p a r t s o f t h e p i
t h e n h o p e f u l l y e x p l o i t i n g som e in f o r m a t i o n t o
p ie c e s a s th e o b je c t s o u g h t.
c la s s ify
th e
r i|
We do n o t d e p a r t i n a n y gi
t h i s s t r a t e g y , s a v e t o i n s i s t t h a t t h e i n f o r m a t i o n ex p lc
e x p lic itly
r a th e r th a n in
t h e s t r u c t u r e o f th e
T h i s d o e s n o t s o u n d u n r e a s o n a b l e : we i n t e n d t o
i t in f a c t i s .
show j u s t how
THE ORIG IN S OF TH IS WORK.
T he w o rk t h a t i s d e s c r i b e d i n t h i s d i s s e r t a t i o n o r i g i n a t e d i n a n i n s t r u ­
m e n ta t io n
p r o b le m .
A
p o p u la r
m illin g
te c h n i q u e
on
many
m in e s
is
r u n - o f - o i n e m i l l i n g , w h e re o r e is m i l l e d a s i t a r r i v e s fro m u n d e r g r o u n d ,
w ith o u t b e in g c ru s h e d .
I t i s d e s i r a b l e t o m i l l t h i s o r e a u to g e n o u s l y .a n d
i t i s b e c o m in g m o re a n d m o re d e s i r a b l e t o I n s t r u m e n t t h e p r o c e s s (Som m er,
1 9 8 5 ).
W ork i s b e i n g d o n e o n i n s t r u m e n t i n g t h e p r o c e s s e s o c c u r r i n g w i t h i n
t h e m i l l d u r i n g m i l l i n g , a n d i t w as f e l t i m p o r t a n t t h a t a n a t t e m p t b e made
t o in s tru m e n t th e s i z e d i s t r i b u t i o n o f th e o r e o f f e r e d t o th e m i l l .
s tr u m e n t s a r e
p r im itiv e
a v a ila b le
a t th is
s ta g e
to
p e rfo rm
(see
th is
a
p a rtic le -s iz e
a re
In ­
e x tr e m e ly
T h is
in v e s tig a te th e f e a s i b i l i t y o f
d is trib u tio n
t e c h n i q u e s , a n d m an ag ed t o
but
t h e A rm co A u to m e tr ic s m a n u a l,1 9 8 4 ) .
a u th o r a n d B e rg e r (1 9 3 4 ) w ere r e q u ir e d t o
b u ild in g
f u n c tio n ,
a n a ly s e r
u s in g - im a g e
p r o d u c e a d e s i g n , w h ic h i s
a n a ly s is
a t p r e s e n t b e in g
im p le m e n te d b y B e r g e r a n d a t h i r d w o r k e r , S m ith .
T he r e s e a r c h r a i s e d many i s s u e s ,
of
p rese n t
im a g e - p r o c e s s i n g
s o l u t i o n te c h n i q u e s a b l e t o
ta k e n
a n i m p o r t a n t o n e b e in g t h e r o b u s t n e s s
m e th o d s ,
u n d er ad v erse c o n d itio n s .
c lu e s t o fin d
n o t h e lp fu l
ro c k s in
( i.e .
a p ic tu re ,
c a llin g
as
t h e p r o b le m r e q u i r e s
h a n d l e c o m p le x p i c t u r e s o f s im p le
a ll
It
is
fo r i ts
o b je c ts ,
n o t p o s s ib le to u se s t r u c t u r a l
a n d s im p ly b i n a r i s i n g t h e p i c t u r e
g re y a re a s
ro ck s,
is
a l l b la i.it, b a c k g r o u n d )
a s a l l r o c k s i n t h e p i c t u r e a r e c o v e r e d w ith , a t h i n l a y e r o f a d s o r b e d f i n e
p a r t i c l e s , a s i s t h e b e l t t o a l a r g e e x t e n t : , an d i t i s q u i t e p o s s i b l e t o
c l a s s i f y w h o le p i c t u r e s i n c o r r e c t l y
T he
te c h n iq u e
d e v e lo p e d ,
as
r e p r e s e n tin g s in g l e
a nd u se d w ith
la rg e ro c k s.
some c o n s i d e r a b l e s u c c e s s , was
s e g m e n t in g t h e p i c t u r e i n t o r e g i o n s , c l a s s i f y i n g t h o s e r e g i o n s , a n d th e n
BACKGROUND
m e rg in g th a m o n t h e b a s i s o £ m e rg in g r u l e s , w h ic h w e re d o vr-Jo p o d ad h o c .
T h e m e rg e r u l e s w e r e b a s e d l a r g e l y o n k n o w le d g e a c q u i r e d b y l o o k in g f it a
ru rm in C " 'c o n v e y o r b e l t w i t h o r e on if. f o r e x a m p le , i t w as r e a l i s e d a t en
e a r ly s ta g e t h a t ,
a s a r e s u l t o f th e p r o c e s s e s a c tin g to p ro d u ce th e o r e ,
few r o c k s w o u ld a p p e a r w i t h h o l e s i n th e m , an d t h a t t h e h o l e s w e r e u n im ­
p o r ta n t.
T h u s, a r e g io n , w h a te v e r i t s
su rro u n d e d b y j u s t
c l a s s i f i c a t i o n , t h a t w as e n t i r e l y
o n e r o c k r e g i o n c o u ld b e m e rg e d t o t h a t
ro c k re g io n
a n d f o r g o t t e n a s a n e n t i t y cm i t s ow n.
T h is t e c h n i q u e w as b a s e d o n a num b er o f a s s u m p tio n s , t h e n s j o r o n e s b e in g
t h a t t h e r o c k s o n t h e b e l t w o u ld b e n i c e o n e s ( i . e . i s n o t e x t r a o r d i n a r i l y
lo n g e n d t h i n o r u n r e a s o n a b l y c u r v e d ) , t h a t a tw o - d im e n s io n a l p i c t u r e o f
a ro ck ly in g o n a v i b r a t i n g
s u r f a c e w o u ld g i v e a g o o d i d e a o f i t s s i z e
(We d o n o t e v e n c o n s i d e r t h e r e a l m e an in g o f s i z e
f o r a ro c k ,
becom e
e x a m p le , M a n d e lb r o t,
e m b ro ile d
i n u n p le a s a n t
is s u e s -
See, fo r
le s t
we
1 9 8 2 .) a n d t h a t s u f f i c i e n t m e rg e r u l e s o f t h e ty p e d e s c r i b e d e x i s t t o f i n d
a re a so n a b le in te r p r e ta t io n
o f th e p ic tu r e .
s u m p tio n s c o u l d b e d e b a t e d :
w h a t i s c e r t a i n i s t h a t t h e s y s te m b a s e d o n
T ie v a l i d i t y
o f th e se a s ­
th e m se em s t o w o rk v e r y w e l l .
From t h i s w o rk so m e im p o r t a n t i d e a s o r i g i n a t e d .
•
F irs tly ,
i t w as s e e n t h a t r e g i o n s a r e a n e x tr e m e ly u s e f u l f e a t u r e t o
e x t r a c t a n d w o rk w i t h , p a r t i c u l a r l y
fo r
r e a l-w o rld
p ic tu r e s
ta k e n
u n d er p o o r c o n d itio n s .
•
S e c o n d ly , i t w as i n f e r r e d t h a t o n e c o u ld w o rk v e r y w e l l w i t h p i c t u r e s
i n tw o d im e n s io n s o n l y .
BACKGROUND
•
T h ird ly ,
it
bticamt- a p p a r e n t
i h a t when o n e i b d e a l i n g w ith a r e g i o n ,
t h e m o st im p o r t a n t i n f o r m ' ^ io n a v a i l a b l e i s
th e n a tu re o f i t s
imme­
d i a t e n e i g h b o u r s , a n d t h a t p p i f t u r e r e p r e s e n t a t i o n t h a t w o u ld c a r r y
t h i s i n f o r m a t i o n c o u ld b e m o st u s e f u l .
a
p ic tu r e
p r o p e r ly
in
th is
w ay ,
r e g i o n ’s n e i g h b o u r s 1 n e i g h b o u r s ,
if
( N o t ic e t h a t i f we r e p r e s e n t
we r e q u i r e i n f o r m a t i o n a b o u t a
th is
is
a ls o o b ta in a b le .
We w i l l
e x p an d on t h i s p o i n t l a t e r . , )
•
F o u r t h l y a n d f i n a l l y , i t w as s o o n s e e n t h a t a s y s te m t h a t a llo w s e x ­
p e r i m e n t a t i o n w i t h e x p l i c i t k n o w le d g e w o u ld b e m o st d e s i r a b l e , a s i f
t h e k n o w le d g e u s e d b y t h e s y s te m w e r e i m p l i c i t , s y s te m m o d i f i c a t i o n s
o r u p d a t e s w o u ld b e c o m p le x .
b e c o m m itte d t o o n e t a s k .
As a r e s u l t t h e s y s te m w o u ld h a v e t o
W ith, a s y s te m w h e r e t h e k n o w le d g e i s
ex­
p l i c i t , h o w e v e r , we may c h a n g e t h e t a r g e t i n g o f t h e s y s te m b y c h a n g in g
t h e k n o w le d g e a v a i l a b l e t o i t .
T h i s w o rk e r t h e n s t a r t e d t o u s e t h e a b o v e r e s u l t s a n d i d e a s fro m t h e w o rk
on p a r t i c l e
s iz e d is tr ib u tio n
s y s te m a b l e t o
h a n d le a l a r g e
a n a ly s is ,
i n an a tte m p t t o b u ild a v is i o n
c l a s s o f s c e n e s u s i n g k n o w le d g e s u p p l i e d
e x p lic itly to i t .
KNOWLEDGE-BASED,SYSTEM S.
I n a s e n s e , a l l s o f t w a r e s y s te m s t h a t s o l v e p r o b le m s a r e k n o w le d g e - b a s e d ;
we u s e t h i s te r m t o r e f e r t o s y s te m s t h a t u s e e x p l i c i t k n o w le d g e i n some
f o rm .
T h i s ty p e o f s y s te m h a s t h e
BACKGROUND
a d v a n ta g e t h a t o n e may im p ro v e t h e
6
p e rf o rm a n c e o f t h e
s y s tw
o r c h a n g e t h e p r o b le m i t
can s o lv e ,
up t o
a
p o i n t , b y im p r o v in g t h e q u a l i t y , o r b y c h a n g in g t h e s u b j e c t o f t h e know ­
le d g e a v a i l a b l e t o
it.
hand , w ill r e q u ir e th e
S y s te m s b a s e d o n i m p l i c i t k n o w le d g e , o n t h e o t h e r
o v e r h e a d o f s y s te m c h a n g e s t o im p ro v e t h e i r p e r ­
fo rm a n c e o r c h a n g e t h e p r o b le m t h a t " t h e y s o l v e .
I m p l i c i t k n o w le d g e s y s ­
te m s make u p f o r t h e i r i n f l e x i b i l i t y b y t h e i r e f f i c i e n c y , a s t h e y d o n o t
r e q u ire
t h e h u g e o v e r h e a d o f k n o w le d g e m an ag e m en t,
e n g i n e e r i n g im a g e p r o c e s s i n g a p p l i c a t i o n s
and a re
as a r e s u l t.
p o p u la r
in
G e n e ra lity w ill
h o w ev e r b e d i f f i c u l t t o a c h i e v e w ith a n i m p l i c i t k n o w le d g e s y s te m .
We c a n s e e t h a t , w h en we h a v e t o a n a l y s e a s c e n e i n a p u r e l y e n g i n e e r i n g
c o n t e x t , i t i s d e s i r a b l e t o em bed t h e d e s c r i p t i o n o f t h e o b j e c t s we a r e
se e k in g
as
s y s te m i s
e f f i c i e n t l y a s p o s s ib le in
lik e ly
to
t h e s y s te m .
W hen, h o w e v e r, o u r
b e f a c e d w i t h a l a r g e nu m b er o f d i f f e r e n t o b j e c t s ,
i t becom es d e s i r a b l e t o b e a b l e t o p r o v i d e t h e s y s te m w i t h a d e s c r i p t i o n
of
e a c h o b j e c t , w i t h o u t h a v in g t o
c h a n g e t h e s y s te m s o f t w a r e .
f i n d e r c a n t h e n , a t a p i n c h , beco m e a n e l e p h a n t - f i n d e r .
t h e h a rd w a r e t e c h n o l o g y may beco m e s u c h t h a t t h i s
A ro ck
A t som e s t a g e ,
may w e l l b e t h e b e s t
a p p r o a c h , fro m a n e n g i n e e r i n g p o i n t o f v ie w : a t p r e s e n t , t h e a p p r o a c h d o e s
n o t p r o m is e r e s u l t s , s a v e a s a t r a i n i n g , d e v e lo p m e n t a n d r e s e a r c h t o o l .
T he b e s t- k n o w n c l a s s o f k n o w le d g e - b a s e d s y s te m i s t h e e x p e r t s y s te m .
B rachm an ( 1 9 8 3 ) p o i n t s o u t , i t
As
i s n o t c o r r e c t t o c a l l a s y s te m t h a t a n a -
ly iA s s c e n e s , a n e x p e r t s y s te m , k n o w le d g e - b a s e d o r r u l e - d r i v e n th o u g h i t
may b e .T h e t a s k i t s e l f d o e s n o t r e q u i r e e x p e r tis e .
c o n s i s t e n t i n t h i s n o m e n c la tu r e .
The l i t e r a t u r e
is not
We u s e t h e te r m k n o w le d g e - b a s e d s y s te m
i n s t e a d , b u t r e g a r d t h e e x p e r t - s y s t e m l i t e r a t u r e a s b e in g r e l e v a n t .
is
b y n o m eans m e r e ly a s e m a n t ic q u i b b l e ;
e a s ily
d e s c r ib e d ,
BACKGROUND
a n d t h e p r o b le m
is
,
e x p e rtis e
in v is io n
is
T h is
not
n o t o n e o f s ., . , p ly d e te r m i n i n g t h e
7
r ig h t r u le s
s o lv e s
an
s p e c ifi
fo r
an e x p c r t - s y s t e m
k e rn e l.
e x p e r t p r o b le m , t h a t i s ,
k n o w le d g e , w h ic h
is
An
g e n e ra lly used in
r e q u i r e o u r s y s te m t o u s e p o o r k n o w led g "
w h ic h e v e r
w ay
it
'c h o o s e s ,
ex p ert
a co m p lex p r o b le m
to
s y s te m
ty p ic a lly
r e q u ir in g h ig h ly
a s p e c if ic
fa sh io n .
We
n a p o o rly -d e fin e d f i e l d ,
s o lv e
a
p r o b le m
in th e
lite ra tu re
th a t
is
not
in
w e ll
u n d e r s t o o d - b y hum an e x p e r t s .
T he
e x p e r t, s y s te m
as
d e s c rib e d
k n o w le d g e b a s e a n d a n i n f e r e n c e e n g i n e .
h a s tw o s e g m e n ts : a
T h e k n o w le d g e b a s e
c o n t a i n s o n ly k n o w le d g e th r o u g h w h ic h t h e .p r o b le m s p o s e d t o
may b e s o l v e d .
T h e i n f e r e n c e e n g in e t h e n e x p l o i t s t h i s k n o w le d g e t o s o l v e
t h e s e p r o b le m s .
our
p u rp o ses.
ty p ic a lly
t h e sy s te m
T h ere a t e v i s i b l e d is a d v a n ta g e s to
The
p r o b le m ; t h e s t r a t e g y
s y s te m
is
cannot
th is
s tr u c tu re
rea so n o u t a s tr a te g y to
t h a t o f t h e s y s t e m 's a u t h o r .
fo r
s o lv e i t s
T y p ic a lly ,
th e
d o m a in o f e x p e r t i s e o f an e x p e r t s y s te m i s l i m i t e d , g e n e r a l l y b y t h e s h e e r
lo g is tic s
o f m a r s h a l l i n g a l a r g e q u a n t i t y o f k n o w le d g e .
i s o f t e n r e s t r i c t e d t o h a v in g i n i t s
i s c o n s i s t e n t w i t h t h e k n o w le d g e b a s e a s a w h o le .
to o
e a s i l y b e c o n s id e re d p u r e ly
s ig n e r to
S u c h a s y s te m
k n o w le d g e b a s e o n l y k n o w le d g e t h a t
T h e k n o w le d g e b a s e c a n
a s a r u l e b a s e , w h ic h c a n c a u s e t h e d e ­
r e s t r i c t t h e p o w e r o f t h e k n o w le d g e .
(One o f t e n
s e e s s y s te m s
d e s c r i b e d a s " f o r w a r d c h a i n i n g " - f o r a p o w e r f u l s y s te m , s u r e l y t h e s y s te m
i t s e l f s h o u l d d e c i d e i n w h ic h d i r e c t i o n i t s h o u ld c h a i n ? )
f o llo w
D a v is
(1 9 8 0 )
in
re fe rrin g
to
We t h e r e f o r e
t h e k n o w le d g e b a s e a s c o n t a i n i n g
k n o w le d g e s o u r c e s ( K S 's ) : i n t h i s c a s e e a c h KS w i l l b e o f a s i m i l a r f o rm a t
t o a n d d i s c h a r g e t h e sam e f u n c t i o n a s , a r u l e i n t h e c o n v e n t i o n a l e x p e r t
s y s te m .
We r e q u i r e o f a k n o w le d g e - b a s e d s y s te m t h a t i t
how i t
s h o u ld m an ag e k n o w le d g e ,
BACKGROUND
th a t i t
b e a b le to
b e a b le t o
rea so n about
c h o o s e b e tw e e n c o n -
B
t r a d i c t o r y K S 's o n t h e g r o u n d s o f o t h e r e v i d e n c e , an d t h a t i t b e l i m i t e d
as l i t t l e
to
add
a s p o s s i b l e i n t h e w ays t h a t i t
t h a t th e
p re s e n te d
by
s id e s te p p e d
p ro to ty p e d e s c rib e d
such
have
r e q u ir e m e n ts :
been,
to
h a n d l e s k n o w le d g e .
ra th e r,
a llo w
We h a s t e n
b e lo w h a s n o t s o lv e d t h e
as
many
is s u e s
as
p r o b le m s
can
t h e d e v e lo p m e n t o f a p r o t o t y p e .
be
The
p r o t o t y p e h a s n o t b e e n d e v e lo p e d a s y e t a n o t h e r e x p e r t system, i t i s i n ­
te n d e d t o d e m o n s tr a te t h e f e a s i b i l i t y
o f k n o w le d g e b a s e d s c e n e a n a l y s i s
a n d o f a new p a ra d ig m w i t h w h ic h we h a v e a t t e m p te d t o s i m p l i f y t h e p r o b le m
o f a n a ly s in g r e a l s c e n e s .
K n o w le d g e -b a se d s y s te m s
lite ra tu re ,
o f w o rk .
a r e b e g i n n i n g t o p e n e t r a t e t h e i m a g e - p r o c e s s in g
an d p r e c e d e n t s d o e x i s t f o r u s i n g t h e s e s y s te m s f o r t h i s ty p e
Nazif a n d L e v i n e ( 2 9 8 4 ) d e s c r i b e a n e x p e r t ( s i c ) s y s te m f o r low
le v e l - i m e g e s e g m e n t a ti o n , w h ic h a p p e a r s t o
a c c o rd in g
s y s te m .
to
th e
p o in t
ra is e d above, i t
g iv e good r e s u l t s ,
is
in
a lth o u g h
f a c t a k n o w le d g e - b a s e d
N iem ann a n d h i s g r o u p ( 1 9 8 5 ) d e s c r i b e a n o t h e r s y s te m w h ic h u s e s
k n o w le d g e t o
in te rp re t r e s u lts
h o w ev o r, t h a t i t
u ses
fro m
a
s e g m e n te d
p ic tu re :
it
a p p e a rs,
i m p l i c i t k n o w le d g e t o se g m e n t t h e p i c t u r e an d e x ­
t r a c t t h e in f o r m a t i o n i t n e e d s f o r i t s a n a l y s i s .
t o u s e k n o w le d g e a t a l l
le v e ls ,
so th a t
We w o u ld l i k e o u r s y s te m
h ig h -le v e l
k n o w le d g e
may
a id
lo w - le v e l a c tio n s .
O ur
s y s te m
is
th o u g h t o f
in
a f a s h i o n d i f f e r e n t fro m t h e c o n v e n t i o n a l
" i n f e r e n c e e n g in e - k n o w le d g e b a s e "
d e v e lo p e d
r e p re s e n ta tio n ;
r a th e r ,
it
a s a n i n t e r p r e t e r o p e r a t i n g o n a p i c t u r e - k n o w le d g e
has been
la n g u a g e .
le te rm " i n t e r p r e t e r " i s u s e d a d v i s e d l y ; d i f f e r e n t r e a s o n i n g f a c i l i t i e s
a re a v a ila b le to
c ility
it
d iffe re n t b its
o f in s tru c tin g
o f k n o w le d g e , a s a t t h i s
s h o u ld m anage k n o w le d g e h a s n o t b e e n
BACKGROUND
s ta g e th e
fa ­
t h e s y s te m , th r o u g h t h e k n o w le d g e b a s e , a s t o how
im p le m e n te d .
T h is
in te rp re te r'
m odel i s
to
a v o id t h e
i m p l i c i t a s s u m p tio n
o f a u th o r - im b e d d e d
r e a s o n in g
te c h n i q u e s i n t h e k n o w le d g e -m a n a g e m e n t s e c t i o n .
SCENE-ANALYSIS PARADIGMS.
T y p i c a l o f t h e s t a t e o f s c e n e a n a l y s i s a t p r e s e n t i s t h e s y s te m p r o p o s e d
by t h i s a u th o r ( F o r s y th , 1984) t o
s o lv e th e
i n s t r u m e n t a t i o n p r o b le m d e ­
s c r i b e d a b o v e , w h ic h e x p l o i t s a s m any p r o b le m c o n s t r a i n t s a s p o s s i b l e t o
p r o d u c e e w o r k a b le d e s i g n .
In
its e lf,
th is
a p p ro a c h i s
n o t in c o rr e c t,
b u t t h e i n v e s tm e n t i n u s i n g a s y s te m b u i l t t o t h i s ty p e o f d e s i g n t o s o l v e
a p r o b le m d i f f e r i n g
even s l i g h t l y
c o n s i d e r a b l e . We w o u ld l i k e
fro m t h e p r o b le m
th e re fo re
f irs t
e n v is a g e d ,
is
to b u ild a sc en e a n a ly s e r t h a t is
a s g e n e r a l a s p o s s i b l e - how d o we d o t h i s ?
We m u st f i r s t d e c i d e i n w h a t fo rm we w is h t o h a n d l e t h e s c e n e . I t i s o b ­
v io u s ly im p o r ta n t t o
be a b le to
e x p lo it th e
s t r u c t u r e o f t h e o b j e c t we
s e e k , y e t we m u s t a l s o d e c i d e how t o h a n d le t h e a p p e a r a n c e o f i t s p a r t s
in th e p ic tu r e .
We may h a n d l e t h e p i c t u r e i n te rm s o f i t s e d g e s , o r o f i t s
edges w i l l
ta tio n :
its
le a d u s t o
re g io n s ,
a fo rm o f l i n e
to
We h a v e c h o s e n t h e l a t t e r ,
r e g io n s .
d r a w in g , o r w ir e - f r a m e
a re p r e s e n ta tio n
and w i l l e x p la in
in
te r m s o f p i c t u r e
and j u s t i f y
Its
rep rese n ­
s u b s e ts .
t h e c h o ic e b o t h
b y r e f e r e n c e t o w o rk i n te r m s o f t h e l i n e - d r a w i n g r e p r e s e n t a t i o n , a n d b y
e x p e r im e n ta l r e s u l t s .
BACKGROUND
T he m a jo r l i n e - d r a w i n g - t y p e p a ra d ig m
a p p e a rs t o b e t h a t o f H arr ( 1 9 7 9 ),
a lth o u g h much o t h e r w o rk h a s b e e n d o n e i n t h i s f i e l d , w h e re p i c t u r e s h a v e
b e e n i n t e r p r e t e d i n te r m s o f - t h e i r e d g e s .
d r a w in g " r a t h e r
lo o s e ly h e re :
w o rk in g i n t h i s
t y p e o f p a ra d ig m .
c y lin d e r
r e p r e s e n ta tio n s
p a ra d ig m i s t h a t o n e i s
m o d e ls
and
if
(We a r e u s i n g t h e te r m
T h u s , t h e te r r a i n c l u d e s g e n e r a l i s e d
th e
lik e .)
r e q u ire d t o
th a t one has b u i l t o f i t .
T h e g r e a t p r o b le m w i t h t h i s
u n d e rs ta n d th e p i c t u r e
i n te r m s , o f
K n o w led g e , n o t o f w h at o b j e c t s
l i k e , b u t o f w h a t t h e y s h o u l d lo o k l i k e i n p i c t u r e s ,
f i c u l t q u e s tio n
" lin e
o n e e x a m in e s a p i c t u r e ' s e d g e s , o n e i s
o f s h a p e c a n b e a v o id e d i n t h i s
d e s c r i b i n g a r o c k e t a s lo o k i n g l i k e
i s used.
lo o k
T he d i f ­
f a s h i o n , f o r e x am p le b y
a cone on to p o f a c y lin d e r , b u t a
s y s te m d e s i g n e d l i k e t h i s i s a t t h e m e rc y o f i t s own m o d e llin g t e c h n i q u e s .
T oo much k n o w le d g e i s d e p lo y e d i n m o d e llin g t h e s c e n e , a n d n o t en o u g h i n
h a n d lin g
it.
T h u s,
k n o w le d g e t o f i n d
a s y s te m b u i l t u n d e r t h i s p a ra d ig m c a n n o t d e p lo y
t h e m o st l i k e l y
p o s itio n
o f a n o b j e c t w h ic h i t
s h o u ld b e t h e r e , b u t w h ic h i t c a n n o t f i n d f o r sh a d o w s .
p ro p o se h e re i s
in te n d e d t o
a llo w
know s
T he p a ra d ig m we
m a r s h a l l i n g o f k n o w le d g e t o
make
it
e ff e c tiv e a t a l l le v e ls .
A n o th e r f e a t u r e w i t h p r e s e n t l i n e - d r a w i n g v i s i o n p a ra d ig m s i s t h e e f f o r t
p u t in to
te c h n i q u e s t o
i n t e r p r e t a tw o - d im e n s io n a l im ag e b y u s i n g c l u e s
t o i t s s h a p e i n t h r e e 1 d im e n s io n s c o n t a i n e d ,
o f th e p i c t u r e .
f o r e x a m p le , i n t h e s h a d in g
I t i s f e l t t h a t t h e s e te c h n i q u e s a r e l i k e l y t o p r o d u c e
m i s l e a d i n g r e s u l t s w hen a p p l i e d t o a r e a l p i c t u r e , p a r t i c u l a r l y on e t a k e n
u nder poo r lig h tin g c o n d itio n s .
i n o n e 's k n o w le d g e b a s e a s t o
sc rib e d ,
I t i s s u r e l y b e t t e r t o h o ld i n f o r m a ti o n
th e
lik e ly
in a p ic tu r e , th a n t o t r y
m odel o f t h e p i c t u r e
i n t h r e e d im e n s io n s ,
p r o b a b ly d e f e c t i v e m o d e l.
BACKGROUND
a p p e a ra n c e o f t h e o b je c ts d e ­
t o obtain a potentially m i s le a d in g
and
th e n
to
in te rp re t
th is
H a r r 's p a ra d ig m
lim its
under i t .
l a b o r i o u s f o r s u c h a s y s te m t o f i n d , a s F r e u d e r ' s ( 1 9 7 6 )
I t is
t h e p ow er o f t h e k n o w le d g e u s e d i n s y s te m s b u i l t
s y s te m c a n , a n o b j e c t i t knows s h o u l d b e p r e s e n t by lenient classification
o t r e g io n s in t h a t o b j e c t 's
m o st l i k e l y
lo c a tio n .
If
h a s b e en p o o rly p ro d u c e d a s a r e s u l t o f b ad l i g h t i n g ,
s y s te m w i l l h a v e d i f f i c u l t y i n m a k in g a n o t h e r t o
of
its
h ig h -le v e l
lo w - le v e l d e c is i o n s .
k n o w le d g e
to
l i n e d r a w in g
f o r e x a m p le , t h e
c o n fo rm t o
le d g e t h a t b eco m e s a v a i l a b l e a t som e l a t e r s t a g e .
p lic a tio n s
th e
o t h e r know ­
I t c a n n o t u s e t h e im ­
im p ro v e
its
h a n d lin g
of
We a r e p r o p o s i n g , t h e n , t h a t p i c t u r e s b e u n d e r s t o o d
i n te rm s o f p i c t u r e s .
L ook a t f i g u r e 1 .
M a rr ( M a r r ,1 9 7 9 ) c la im s t h a t t h i s p i c t u r e i s p e r c e i v e d
" i n t e r n s of" v e r y p a r t i c u l a r t h r e e - d i m e n s i o n a l s h a p e s , som e f a m i l ­
i a r , som e l e s s s o . "
This a u t h o r d i s p u t e s h i s
to
p ro v id e c lu e s a s t o
c la i m ;
s u r e l y t h e nam e of t h e p i c t u r e
is
u sed
th e id e n t i t y o f th o s e o b je c ts n o t c le a r l y c l a s s ­
i f i a b l e , a n d th o s e t h a t a re u n d e rs to o d w ith o u t t h e s e c lu e s , a re u n d e rsto o d
a s tw o - d im e n s io n a l b l o t c h e s .
to rily
c la s s ifie d ,
d im e n s io n s .
a ll
W h ile i t
T h en , once e v e r y th in g h a s been
o b je c ts
s a tis fa c ­
i n h e r i t fro m t h e v ie w e r a s h a p e i n t h r e e
u n f a ir to
c ritic is e
t h e p a ra d ig m o n t h e b a s i s
o f a c o m p le x p i e c e o f a r t w o r k , w h e re b a c k g r o u n d k n o w le d g e b eco m es im p o r ­
t a n t b y t h e n a t u r e o f t h e p i c t u r e , t h e d a . ', , ) h o l d s t h a t t h r e e - d i m e n s i o n a l
shape is o f l i t t l e
im p o r ta n c e i n t h e e a r l y s t a g e s o f c l a s s i f y i n g a s c e n e .
We c la im t h a t s h a p e i s o f t e n i n h e r i t e d b y o b j e c t s i n a p i c t u r e a f t e r th e y
have
been
c la s s ifie d ;
th is
e ffe c t is
c o n s id e r s p i c t u r e s o f p o o r q u a l i t y .
p a rtic u la rly
In th e ro ck
n o t i c e a b l e when o n e
p ic tu r e s
sh o w n ,
th e 1
r e a d e r s h o u ld n o t i c e t h a t t h e r o c k s a c q u i r e s h a p e o n ly a f t e r h e h a s p u z -
BACKGROUND
T he
rite
of
spring.
z l e d o u t w h ere t h e y a r e , n o t w h ile h e is. d o in g s o ; th u s in f o r m a t i o n a b o u t
t h e tw o - d im e n s io n a l a p p e a r a n c e o f o b j e c t s i s b e in g e x p l o i t e d .
T he am o u n t o f i n f o r m a t i o n
in v o lv e d
l u s t r a t e d by t h e ro c k p i c t u r e s .
in u n d e rs ta n d in g sc e n e s
is
w e ll i l ­
H o s t p e o p l e who h a v e s e e n t h e s e p i c t u r e s ,
d i d n o t know w h a t t h e y r e p r e s e n t e d u n t i l th e y w e r e t o l d .
E v en t h e n th e y
w e re n o t s u r e w h e r e a b o u ts i n t h e p i c t u r e s t h e r o c k s w e r e , a n d c o u ld I n ­
te rp re t
th e
p ic tu r e s
p l a i n e d t o th e m .
c o r r e c t l y o n ly a f t e r h a v in g s e v e r a l p i c t u r e s
From t h i s s t a g e o n , t h e y w e r e a b l e t o
p ic t u r e s f o r th e m s e lv e s .
ex­
in te r p r e t th e se
T h u s t h e p r o c e s s i s l e a r n t , an d i n s t a g e s .
O n ly
o n c e t h e s e s t a g e s h a v e b e e n t r a v e r s e d , :Ls t h e p i c t u r e u n d e r s t o o d i n te r m s
o f th e th r e e - d im e n s io n a l sh a p e o f th e o b je c ts r e p r e s e n te d .
We h a v e
la b o u r e d
th e
r e l a t i v e u n im p o r ta n c e o f t h r e e - d i m e n s i o n a l s h a p e
in f o r m a tio n in th e e a r l y s ta g e s o f sc e n e a n a ly s i s , b e c a u se
d e s c rib e d
h e re
ig n o re s t h i s
i n f o r m a t i o n c o m p le t e ly .
th e
s y s te m
We f e e l t h a t t h e
r e s u l t s o b t a i n e d i n t h e p a r t i c l e - s i z e d i s t r i b u t i o n a n a l y s i s w o rk j u s t i f y
t h e tis e o f t h e te c h n i q u e s d e s c r i b e d ,
p lo ita tio n
a lth o u g h t h e y make t h e
o f t h r e e - d i m e n s i o n a l s h a p e in f o r m a t i o n d i f f i c u l t .
la te r
ex­
The r e l a ­
t i v e u n im p o r ta n c e o f t h i s i n f o r m a t i o n m ean s t h a t t h e i r i n a b i l i t y t o h a n d le
t h i s in f o r m a t i o n d o e s n o t m i l i t a t e a g a i n s t th e m .
IMAGE REPRESENTATION.
I t i s i m p o r ta n t t o c o n s i d e r how im a g es may b e r e p r e s e n t e d , a s t h e c h o ic e
o f im ag e r e p r e s e n t a t i o n w i l l o f t e n i n f l u e n c e t h e a b i l i t i e s o f t’h e s y s te m
BACKGROUND
in q u e s tio n .
We. w is h t o
re p re s e n t th e
p o s s i b l e , w h ile c o n s e r v in g
th e p ic tu r e
in
a fa sh io n
im a g e i n
a s p o w e r f u l a fo rm
a s much s p a c e a s p o s s i b l e .
c o m p a ti b le w i t h t h e
as
We w is h t o s t o r e
s y s te m we a r e d e s c r i b i n g ,
t h a t i s , we w is h t o h o l d in f o r m a t i o n p e r t a i n i n g t o t h e r e g i o n s w i t h i n t h e
p i c t u r e i n a s t r a c t a b l e a fo rm a s p o s s i b l e .
For
th is
p u rp o se ,
it
m u st
be
a ssu m ed
t r u s t w o r t h y , t h a t i s , t h a t i t w i l l n o t b<
to ask i t to re v is e i t s d e c is io n s .
l a t e r s ta g e
th a t each
T h i s i s t h e saim
p r im it iv e i s r e p re s e n te d by a t l e a s t one r e g io n in
p ic tu re .
a n d t h a t n o tw o p r i m i t i v e s s h a r e a r e g i o n , w h e re
a p rim itiv e
s u n d e rs to o d by a
(T hat i s ,
s tr u c tu re to fin d a p r im itiv e :
i t s own r i g h t . )
it
is
we d'
in s p e c t
a p ie c e o f p i c t u r e t h a t s ta n d s i n
S o , o n c e we h a v e se g m e n te d a p i c t u r e , we may d i s c a r d t h e
p i c t u r e i t s e l f i n f a v o u r o f a r e p r e s e n t a t i o n o f t h e se g m e n te d p i c t u r e t h a t
is
m o re c o m p a c t t h a n t h e p i c t u r e
we r e q u i r e .
its e lf,
if
it
re ta in s
t h e in f o r m a t i o n
T h is i s s u e
T h e q u e s t i o n i s , w h a t d o we w is h t o r e p r e s e n t ?
is
n o t y e t fu lly
s e ttle d ,
a s th e
s y s te m
its e lf
is
s till
a
p r o t o t y p e ; h o w e v e r, we ^ h a l l d e s c r i b e w h at w as r e p r e s e n t e d a n d how , an d
we c la im t h a t t h e s y s te m a llo w s e x p a n s io n o f t h e r e p r e s e n t a t i o n .
As h a s
b e e n s e e n a b o v e , t h e m o st i m p o r ta n t i n f o r m a t i o n t h a t o n e may h a v e a b o u t
a r e g io n a p p e a rs t o
We w is h a l s o
to
be th e
lis t,
h e ld
i n som e f o rm , o f i t s n e i g h b o u r s ,
h o ld some r e p r e s e n t a t i o n
l e v e l , i t s t e x t u r e , and i t s
of its
s iz e ,
its
a v erag e g rey
s h a p e i n tw o - d im e n s io n s .
I t c a n b e s t a t e d c l e a r l y a t t h i s p o i n t t h a t t h e i s s u e o f tw o - d im e n s io n a l
s h a p e h a s b e e n l a r g e l y a v o id e d ; i n i t s p r e s e n t s t a t e , t h e p r o t o t y p e sy ste m .
n e ith e r needs n o r u se s
th e
in f o r m a tio n .
becom e an i m p o r ta n t i s s u e , a n d t h a t i t
o f p ic tu r e s
a t a h ig h
It
is
is
a c c e p te d
c r u c ia l to
th a t th is w ill
th e
r e p re s e n ta tio n
le v e l to fin d a r e p r e s e n ta tio n o f shape t h a t w i l l
a llo w o n e t o d e c i d e w h a t t h e s h a p e w i l l b e o f a r e g i o n c r e a t e d b y m e rg in g
tw o o t h e r r e g i o n s o f know n s h a p e , w i t h o u t r e f e r e n c e t o t h e p i c t u r e i t s e l f .
T he i s s u e h a s s im p ly c a u s e d
i n s u f f i c i e n t p r o b le m s u p t o
w a r r a n t m o re t h a n b r i e f a t t e n t i o n ;
ta tio n
th is
s ta g e
to
i t i s f e l t t h a t th e r e g io n r e p r e s e n ­
a d o p te d h a s s u f f i c i e n t p o w e r , w i t h
some
a tte n tio n ,
to
a llo w
a
s o lu M o n o f t h i s p r o b le m w i t h o u t m o d i f i c a t i o n t o t h e r e p r e s e n t a t i o n .
T he i s s u e o f r e p r e s e n t i n g t h e
e v er,
re c e iv e d
p a rtic le -s iz e
much
r e g i o n s t r u c t u r e o f t h e p i c t u r e h a s , how ­
a tte n tio n ,
d is trib u tio n
d a tin g
fro m
a n a l y s e r w o rk .
th e
e a r ly
days
At th a t s ta g e ,
t a t i o n w as r e q u i r e d t h a t i n d i c a t e d how a r e g i o n r e l a t e d
to
of
th e
a rep rese n ­
its
o u ts id e
n e i g h b o u r s , t h e i n s i d e n e ig h b o u r s b e in g ig n o r e d f o r r e a s o n s we s h a l l d e ­
ta il
la te r .
R o s e n f e ld
The
(1 9 8 2 ),
in tra c ta b ility .
c h a in - c o d e
w as
a p p ro a c h ,
u n a p p e a lin g
In th e
on
ro ck p ic tu r e s
as d e ta ile d
th e
g ro u n d s
o b ta in e d ,
lo n g b o u n d a r i e s , w i t h many c h a n g e s o f d i r e c t i o n
n e ig h b o u r s r e p r e s e n t e d .
i n , f o r e x a m p le ,
of
its
p o te n tia l
r e g i o n s te n d e d t o h a v e
i n th e m , b u t
w ith
few
T h u s , a v a s t c h a i n c o d e c o u ld b e o b t a i n e d w ith
m uch l a r g e l y - r e d u n d a n t i n f o r m a t i o n .
T h e n e ig h b o u r - c o d e
r e p r e s e n ta tio n
w as th e n s u g g e s te d b y B e r g e r (1 9 6 4 ( a ) ) a n d th e n r e f i n e d b y t h i s
a u th o r.
T h is r e p r e s e n ta tio n w i l l be d e s c r ib e d i n d e t a i l i n th e body o f th e d i s ­
s e rta tio n .
T he s t a n d a r d d e v i a t i o n o f t h e g r e y l e v e l o v e r t h e r e g i o n h a s b e e n c h o sen ,
to re p re s e n t te x tu r e .
T h is c h o ic e i s o p e n t o d i s c u s s i o n :
n o t t h e o n l y way i n w h ic h o n e may r e p r e s e n t t e x t u r e .
i t i s c e r ta in ly
H o w ev er, we h a v e
a g a i n n o t a d d r e s s e d t h i s i s s u e i n d e t a i l , a s t h e m a in t h r u s t o f t h e w o rk
BACKGROUND
15
w as i n te n d e d t o
e s ta b lis h
th e
f e a s i b i l i t y an d u s e f u l n e s s o f t h e ty p e o f
s y s te m d e s c r i b e d .
KNOWLEDGE REPRESENTATION
R e p r e s e n t i n g k n o w le d g e i s
n o t t h e sam e p r o b le m
d a t a : h o w e v e r, we r e q u i r e
fro m o u r s o l u t i o n t o
sam e c h a r a c t e r i s t i c s a s we r e q u i r e
s im p le , t r a c t a b l e
te c h n iq u e
s a c r if ic e d fo r s im p lic ity .
is
fro m t h e
r e q u ire d
as r e p r e s e n tin g
p ic tu r e
t h e f o r m e r p r o b le m , t h e
s o lu tio n to
th e
la tte r.
A
a t t h i s s t a g e , an d p o w e r may b e
T h u s , a lth o u g h t h e
is s u e w i l l be d is c u s s e d ,
o u r k n o w le d g e f o r m a t w i l l n o t i n a n y way r e p r e s e n t t h e s t a t e o f t h e a r t .
T h e r e a r e s e v e r a l k n o w le d g e r e p r e s e n t a t i o n s p r o p o s e d a t p r e s e n t : t h e m a in
th e m e s seem t o b e s e m a n t ic n e t s , f r a m e s , r u l e s a n d t h e p r e d i c a t e c a l c u l u s ,
w h ile
o th e r
r e p r e s e n ta tio n s
have
been
p ro p o se d .
(S e e
fo r
e x a m p le ,
C h a r n ia k ( 1 9 8 1 ) , a p a p e r w h ic h p r o p o s e s a k n o w le d g e r e p r e s e n t a t i o n
h a s t h e f l a v o u r b o t h o f f ra m e s a n d t h e p r e d i c a t e c a l c u l u s . )
th a t
C o n s i d e r in g
a p i e c e o f k n o w le d g e a s a r u l e , a n d t h i n k i n g o f i t s o l e l y i n t h o s e te rm s
a p p e a rs t o
r e s t r i c t o n e 's p e r c e p t i o n o f t h e p o t e n t i a l o f t h e k n o w le d g e ,
a s i t i s to o e a sy t o g e t d i s t r a c t e d b y th e im p lic a tio n s o f th e " i f - th e n "
c o n s tr u c t.
is
T h u s , a l t h o u g h t h e k n o w le d g e r e p r e s e n t a t i o n we h a v e a d o p te d
in a r u l e - l i k e
f o r m a t, we h a v e a v o id e d s a y i n g " i f
th is
th e n t h a t" .
T he r e p r e s e n t a t i o n i s r a t h e r a s a " t h i s r e l a t e s t o t h a t " fo rm .
B a r r an d F e ig e n b a u m ( 1 9 6 1 ) p r e s e n t a d i s c u s s i o n o f t h e s e te c h n i q u e s w h ic h
we s h a l l s u m m a rise h e r e . . L o g ic a l r e p r e s e n t a t i o n s
BACKGROUND
such as th e p r e d ic a te
tu r a l re p r e s e n ta tio n fo r c e r ta i n n o tio n s .
H o w ev e r, i t
th a t lo g ic s need to
th is • re n d e rs
p u rp o se a t l e a s t ,
b e n o n - m o n o to n ic * 'a n d
a s tr u c tu re
i n w h ic h k n o w le d g e i s
l i n k s o f d i f f e r e n t ty p e s b e tw e e n n o d e s .
a tio n s , o b je c ts
th e m ,
fo r
our
in tr a c ta b le .
A s e m a n t ic n e tw o rk i s
tw e e n th e m .
i s o f t e n th e c ase
N odes u s u a l l y
r e p r e s e n te d by
re p re se n t s it u ­
o r c o n c e p ts , and th e lin k s th e r e l a t i o n s t h a t e x i s t b e ­
S e m a n tic n e t w o r k s , a n d a n e l a b o r a t i o n o f t h e n e tw o r k i d e a
i n t o a s t r u c t u r e , t h e f r a m e , a r e t h e s u b j e c t s o f much r e s e a r c h .
T h e r e p r e s e n t a t i o n a l s t r u c t u r e w i t h w h ic h we a r e c o n c e r n e d , t h e r u l e , h a s
b e e n m uch i n v o g u e r e c e n t l y , w i t h t h e s u c c e s s o f e x p e r t - s y s t e m t e c h n o lo g y
in
c o m m e r c ia l a p p l i c a t i o n s .
The id e a a c t i n g h e re
k n o w le d g e may b e e x p r e s s e d i n a n " I F
T h is
re p r e s e n ta tio n
has
k n o w le d g e , a n d se em s t o
o f k n o w le d g e r e q u i r e d t o
is
been used v e ry s u c c e s s fu lly
le n d i t s e l f
s o lv e t h i s
t a g e s : p r e s e n t e x p e r t s y s te m s
th a t
a
p ie c e
of
( c o n d i t i o n s ) THEN ( a c t i o n s ) " fo rm .
f o r e x p e r t d o m a in
v 'r y w e ll t o r e p r e s e n tin g
k in d o f p r o b le m .
th e
ty p e
I t h a s d is a d v a n ­
f a c e p o t e n t i a l p r o b le m s w i t h
s a tu r a tio n ,
w h e re t h e r u l e b a s e b eco m e s s o l a r g e t h a t t h e s y s te m h a s r e a l d i f f i c u l t y
g e t t i n g a n y th in g done.
R ecent
ru le s to
1
w o rk h a s show n t h a t s y s te m s may b e b u i l t t h a t c h a n g e o r u p d a te
a llo w th e m t o h a n d l e s i t u a t i o n s
t h a t t h e y h a v e n o t , p r e v i 6 u s ly
A m o n o to n ic l o g i c i s a l o g i c a l s y s te m i n w h ic h , i f
i t i s p o s s ib l e to
d e d u c e a th e o re m w i t h a s e t o f a s s e r t i o n s , a d d i t i o n s t o t h a t s e t w i l l
n o t l a t e r n e g a t e t h e th e o re m .
BACKGROUND
'
17
e n c o u n te r e d ( L e n a t , 1 9 8 0 ) .
B rachm an c t a l
( 3 9 8 3 J q u o te H a r t fls s a y in g in
a 1980 p a p e r t h a t c u r r e n t e x p e r t s y s te m s t y p i c a l l y show u p b a d ly , a s th e y
a r e u n a b le t o
le d g e i s
r e c o g n i s e o r d e a l w i t h p r o b le m s f o r w h ic h t h e i r own know ­
i n a p p l i c a b l e o r i n s u f f i c i e n t : t h i s s i t u a t i o n w o u ld a p p e a r t o b e
c h a n g in g , b u t t h e c r i t i c i s m
is s t i l l
la rg e ly
v a lid .
T h e s e p r o b le m s a r e
t o an e x t e n t i n h e r i t e d fro m t h e r u l e r e p r e s e n t a t i o n o f k n o w le d g e .
We h a v e r e v ie w e d h e r e t h e m a jo r i d e a s i n k n o w le d g e r e p r e s e n t a t i o n .
The
c r i t e r i o n b y w h ic h a te c h n i q u e w as c h o s e n f o r t h e p r o t o t y p e s y s te m was
a g a i n s i m p l i c i t y o f i m p l e m e n ta tio n ,
r e c a l l i n g t h a t a r e p r e s e n t a t i o n was
d e s i r e d , l e s s f o r t h e m a in te n a n c e o f p r o b le m k n o w le d g e , th a n f o r p r o v i d i n g
a s im p le e x p l i c i t k n o w le d g e f a c i l i t y
s e n t a t i o n w as c h o s e n f o r i t s
i n t h e s y s te m .
A ru le -ty p e
rep re­
p o t e n t i a l s i m p l i c i t y o f im p l e m e n ta tio n
in
FORTRAN ( w h ic h la n g u a g e c h o ic e w i l l b e e x p l a i n e d ) a n d b e c a u s e i t a p p e a r e d
t h a t u p d a t i n g t h e t h e k n o w le d g e b a s e c o u ld b e s i m p l i f i e d b y u s i n g t h i s
r e p r e s e n ta tio n .
I t w i l l b e se e n l a t e r t h a t th e r u le s have been u se d in
d i f f e r e n t w a y s , d e p e n d in g o n t h e i r te r m s o f r e f e r e n c e ; t h e r e i s a s t r o n g
sim ila rity
to
an in te r p re te d
la n g u a g e w h e re s t a t e m e n t o r d e r i s
n o t im ­
p o r ta n t.
KNOWLEDGE MANIPULATION.
A n o th e r im p o r t a n t q u e s t i o n w h ic h a r i s e s i s ,
th a t is
"How we m a n ip u la te k n o w le d g e
i n a c c u r a t e : how d o we r e a c h a d e c i s i o n b a s e d o n p o o r e v id e n c e ? "
T h is i s a g a i n a r e s e a r c h i s s u e i n i t s e l f , a n d t h e s y s te m we h a v e a d o p te d
d o e s n o t c la im t o b e o p t i m a l .
BACKGROUND
As t h e te c h n i q u e s d e v e lo p e d s o f a r a p p e a r
18
t o b e i n s largo m e a s u r e o n i p i r j c a ] , a n d a s t l i o r e appears t o b o n o s t r o n g
u n i f y i n g i d e a a t p r e s e n t i n t h i s f i e l d , we h a v e a g a i n c h o s e n t h e s i m p l e s t
te c h n iq u e t h a t w i l l m eet o u r p r e s e n t n eed s in t h i s m a tte r.
T he p r o b le m o f r e a c h i n g a d e c i s i o n b a s e d o n p o o r e v id e n c e i s a c o n s i d e r ­
a b l e o n e , a n d a l t h o u g h n o t a n im m e d ia te p r o b le m i n t h e c a s e o f t h e p r e s e n t
p r o t o t y p e , w h ic h w i l l s im p ly make t h e w ro n g d e c i s i o n w hen f a c e d w i t h p o o r
e v id e n c e ,
it
is one t h a t r e q u i r e s a d d r e s s i n g , a s t h e s y s te m i s in t e n d e d
t o b e a p r o to ty p e o f a d e v ic e t h a t w i l l o p e ra te on r e a l p i c t u r e s ,
h ence w i l l r e q u ir e
a c e rta in
and
d e g r e e o f n o i s e im m u n ity .W e w i l l t h e r e f o r e
b r i e f l y s u m m a rise t h e te c h n i q u e s a v a i l a b l e .
One may u s e a B a y e s ia n t e c h n i q u e , w h e re a c o m b in a tio n o f a p r i o r i p r o b a ­
b ilitie s
S h o rtliffe
le a d s
et
to
al
a
p ro b a b ility
(1 9 8 4 )
q u o te
on
w h ic h
a
Duda
as
u s in g
PROSPECTOR s y s te m , w i t h c o n s i d e r a b l e s u c c e s s .
d e c is io n
th e se
may
be
m e th o d s
b ased.
in
th e
T h e p r o b le m w i t h B a 'y e sia n
te c h n i q u e s if f t h a t t h e a, p r i o r i p r o b a b i l i t i e s a r e n o t a lw a y s e a s y t o f i n d ,
a n d t h e t e c h n i q u e se em s t o b e u n p o p u la r a s a r e s u l t .
D uB ois e t a l
( 1 9 8 0 ) p r e s e n t Z a d e h 's c o n c e p t o f t h e
s e t ’ s m em bers a r e
fu zz y s e t ,
w h e re
a
a s s i g n e d a d e g r e e o f m e m b e rsh ip , an d w h e re f u n c t i o n s
may b e d e f i n e d t o d e s c r i b e t h e d e g r e e o f m e m b ersh ip o f an o b j e c t i n t h e
u n io n
o r in te r s e c tio n
o f th e s e t s .
The t h e o ry i s
de ep and r a t h e r a p ­
p e a l i n g , a s f u z z y f u n c t i o n s and f u z ^ y n u m bers may b e d e f i n e d .
T h ere i s ,
h o w e v e r, fit l e a s t o n e c o n s i d e r a b l e p r o b le m t o b e c o n s id e r e d i n u s i n g t h i s
t h e o r y i n an im p le m e n te d s y s te m .
C o n s i d e r t h e s e t o f s t u d e n t s i n a c l a s s ; we w is h t o d e f i n e f o r e a c h s t u ­
d e n t h i s d e g r e e o f m e m b ersh ip i n t h e s e t o f t a l l
BACKGROUND
p e o p le .
A se v e n -fo o t
.
19
s t u d e n t i s a member t o d e g re ct 1 ( t h e s c a l e it. c o n v e n t i o n a l l y
t o w h at d e g re e i s
a s i x - f o o t s t u d e n t a member ?
w h at t h e m e m b ersh ip f u n c t i o n s h o u ld lo o k l i k e .
0 - 1) ; b u t
T h e p r o b le m i s s im p ly
T h e r e a r e o t h e r p r o b le m s :
g e n e r a l l y , w i t h t h e f u n c t i o n s d e s c r i b e d i n D u B o is e t a l ( 1 9 8 0 ) , t h e t r u t h
v a l u e o f t h e p r o p o s i t i o n d o e s n o t n e c e s s a r i l y i n c r e a s e a s e v id e n c e m o u n ts ,
a nd a " s t r o n g " p i e c e o f e v id e n c e may b e o u tw e ig h e d b y a w eak o n e .
does n o t e a s il y
T h is
l e a d t o a g o o d ,m o d e l o f human e v id e n c e c o m b in a t io n .
F o r t h e MYCIN s y s te m , S h o s t l i f f e .
t o w o rk r a t h e r w e l l .
( 1 9 6 3 ) d e v e lo p e d a s y s te m t h a t a p p e a r s
I t d o e s h o w e v e r, a p p e a r t o
have
been
d e v e lo p e d
r a t h e r e m p i r i c a l l y w ith t h e MYCIN i s s u e s i n m in d , a n d a s s u c h w o u ld n o t
s u i t o u r p u r p o s e , a s i t c a n n o t b e l i f t e d s t r a i g h t c u t o f t h e MYCIN c o n t e x t
and u se d .
( S h o r t l i f f e ( 1 9 6 3 ) p o i n t s o u t w e a k n e s s e s i n t h e s y s te m , w i t h
t h e p r o v i s o t h a t MYCIN i t s e l f w o rk e d v e r y w e l l . )
As i n a l l t h e o t h e r f i e l d s we h a v e d i s c u s s e d , we w is h e d f o r o u r p r o t o t y p e
t h e s i m p l e s t p o s s i b l e t e c h n i q u e t h a t w o u ld s e r v e o u r p u r p o s e s .
fu z z y s e t r e p r e s e n t a t i o n w as u s e d ,
T hus t h e
b o th b e c a u s e i t d e f a u l t s i n t h e c a s e
o f d e f i n i t e e v id e n c e t o t h e B o o le a n m o d e l, a n d b e c a u s e we d i d n o t r e q u i r e
e x t e n s i v e w o rk w i t h p o o r e v id e n c e .
REASONING ABOUT,'CONTROL.
We h a v e s t a t e d how i m p o r ta n t i t i s t h a t a k n o w le d g e - b a s e d s y s te m b e a b l e
t o r e a s o n a b o u t how i t w i l l u s e i t s
seem s t o
k n o w le d g e .
r e f l e c t no c l e a r .tr e n d s a t t h i s
BACKGROUND
s ta g e ,
A lth o u g h t h e
th is
lite ra tu re
a u th o r f e e l s t h a t
t h e s o l u t i o n t o t h i s p r o b le m l i e s b o th i n u s i n g th r. te c h n i q u e o f d i v e r t i n g
th e i n t e r p r e t e r 's a t te n t io n
t o c e r t a i n K S 's a t t h e e x p e n s e o f o t h e r s , and
i n p r o v i d i n g t h e i n t e r p r e t e r w i t h d i f f e r e n t te c h n i q u e s o f k n o w le d g e e x ­
p lo ita tio n ,
and
a llo w in g
t h e k n o w le d g e b a s e t o
f o r c e s w itc h e s b e tw ee n
th e m , g e n e r a l l y a s a r e s u l t o f t h e c o n t e x t i n w h ic h k n o w le d g e a p p e a r s .
F o r e x a m p le , t h e f i r s t t e c h n i q u e w o u ld r e s u l t i n a s c e n e a n a l y s i s s y s te m
w h ic h w hen a s k e d t o f i n d a n e l e p h a n t i n a s c e n e , w o u ld p r e v e n t k n o w le d g e
a b o u t d i f f e r e n t a n im a l s fro m r e a c h i n g
d is c u s s e s
D a v is
(1 9 8 0 )
e t e c h n i q u e w h e re p l a u s i b l y - u s e f u l k n o w le d g e s o u r c e s
th e
in te r p re te r .
(P K S 's)
a r e f o u n d , b y f i r s t r e j e c t i n g t h o s e K S 's w h ic h a r e d e f i n i t e l y n o t u s e f u l ,
and th e n o r d e r in g by u t i l i t y
k n o w le d g e - d r i v e n .
It
th e r e s t .
T h e s e p r o c e s s e s a r e th e m s e lv e s
i s n o t e a s y , h o w e v e r, t o s e e a r e l i a b l e te c h n i q u e
w h ereb y k n o w le d g e may b e s u p p r e s s e d w i t h o u t l o s i n g
w hose r.e ’ - v m e e i s n o t im m e d ia te ly o b v io u s .
r e le v a n t
k n o w le d g e ,
I n o u r p r o t o t y p e s y s te m a l l
p i e c e s o f k n o w le d g e a r e b r o u g h t t o t h e a t t e n t i o n o f t h e i n t e r p r e t e r , an d
th o s e w h ic h d o n o t f u l f i l a r e l e v a n c e c r i t e r i o n a r e ig n o r e d .
T h is i s a
p r i m i t i v e i m p l e m e n ta tio n o f t h e a b o v e t e c h n i q u e .
T he s e c o n d t e c h n i q u e i s i l l u s t r a t e d i n o u r p r o t o t y p e w hen we c o n s i d e r t h e
k n o w le d g e t h a t d e s c r i b e s s t r u c t u r e .
I f a KS r e l e v a n t t o
th e s tr u c t u r a l
d e s c r i p t i o n we a r e s e e k i n g i s lo a d e d i n t o t h e i n t e r p r e t e r , i t i s in s p e c t e d
t o a s c e r t a i n w h a t o t h e r k n o w le d g e may b e r e l e v a n t , an d t h e n h e l d f o r f u tu r .P r e fe re n c e ;
la te r ,
w hen we a r e a c t u a l l y
l o o k in g
fo r th a t s tr u c tu re
i n t h e s c e n e , t h e k n o w le d g e w i l l b e e v a l u a t e d a s a r u l e t o i n d i c a t e w h at
v a lu e we a s s o c i a t e w i t h a g iv e n s e t o f r e g i o n s b e lo n g in g t o
BACKGROUND
th a t s tr u c -
IMPLEMENTATION ISSUES.
T he
s y s te m
d e s c rib e d
m i n i - c o m p u te r .
fa m ilia rity
a v a ila b le
w as
im p le m e n te d
i n HP FORTRAN-77, on a n H P-1000
FORTRAN w as u s e d b e c a u s e o f t h e
w ith
w e re
th e
an
la n g u a g e
u n te s te d
an d
su b set
because
of
a u t h o r 's
th e
L IS P ,
o n ly
and
c o n s id e ra b le
o th e r
PASCAL.
la n g u a g e s
To
a v o id
i n t e r p r e t e r - t e s t i n g i s s u e s , a n d b e c a u s e t h e PASCAL c o m p ile r a v a i l a b l e a t
t h e tim e w as e x tr e m e ly s lo w , FORTRAN w as t h e o b v io u s c h o i c e .
The code
w as d e s i g n e d , w r i t t e n , a n d t e s t e d fro m May t o A u g u s t, 1 9 6 5 , i n c l u s i v e .
As t h e H P -1000 u s e d h a s a maximum p a r t i t i o n - s i z e
it
becam e n e c e s s a r y t o
(c o d e an d d a t a )
o f 32K
se g m e n t t h e c o d e , w h ic h h a d n o t o r i g i n a l l y b e e n
w r i t t e n w i t h t h i s i n m in d .
T h e w o rd se g m e n t w i l l r e f e r b e lo w t o t h e c o d e
s e g m e n t s , w h ic h u n f o r t u n a t e l y c o r r e s p o n d t o t h e s y s te m b l o c k , l e a d i n g t o
p o t e n t i a l c o n f u s i o n u n l e s s o n e 's n o m e n c la tu r e i s c a r e f u l .
CONCLUSION.
We h a v e i n t r o d u c e d t h e i d e a s t h a t a p p e a r i n t h e r e s t o f t h i s d i s s e r t a t i o n
and a tte m p te d t o
d e p a rts
show t h e i r o r i g i n s .
T h e w o rk t h a t i s
d e s c r i b e d b elo w
i n many w ay s fro m t h e m a in s tr e a m o f c u r r e n t v i s i o n w o rk , b u t i s
r o o te d in r e s p e c t a b le c o n c e p ts .
T he p r o t o t y p e b u i l t
is
s till
a f ra g ile
c u r i o s i t y , r a t h s r th ,m an e n g i n e e r i n g s o l u t i o n , y e t i t r e p r e s e n t s a n a t ­
te m p t Vf'
a n d tw o „
BACKGROUND
■ 5})? i d e a s p r e s e n t e d , w i t h te c h n i q u e s d e v e lo p e d b y t h i s a u th o r
w o r k e r s , t o p r o d u c e a s y s te m t f i a t c a n s e e .
NEIGHBOUR CODES.
(Some o f t h e w o rk d e s c r i b e d h e r e w as s u b m i tte d
in
a re s e a rc h r e p o r t to
t h e W its D e p a rtm e n t o f E l e c t r i c a l E n g in e e r in g f o r a B . S c . ( E l e c . E n g . ) , i n
W hich t h e
c o n c e p t o f n e ig h b o u r c o d e s ,
a n d t h e w ay
th a t
m e rg e d w h en n o e n c l o s e d r e g i o n s e x i s t e d , w e re d e s c r i b e d .
th e y
c o u ld
be
T h e a lg o r i t h m
t o d e a l w i t h t h i s s i t u a t i o n w as d e v e lo p e d a f t e r t h i s r e p o r t w as s u b m i t t e d ,
and t h e m a t e r i a l i s p r e s e n t e d a s a w h o le f o r s i m p l i c i t y 's s o k e . )
We h a v e s a i d a b o v e t h a t w h en o n e c o n s i d e r s a p i c t u r e i n te r m s o f r e g i o n s ,
t h e i r n e i g h b o u r s ’ q u a l i t i e s b ecom e t h e m o st im p o r t a n t i n f o r m a t i o n
we
can p r o v id e .in
its e lf .
a r e p r e s e n ta tio n
H o w ev e r, i t
a t a h ig h e r le v e l
a l s o b ecam e a p p a r e n t t h a t r e p r e s e n t a t i o n s
th e c h a in co d e w ere n o t i d e a l .
th a t
-b an th e p ic tu r e
su c h as
T h e s e f a c t s s p u r r e d t h e d e v e lo p m e n t o f
t h e i d e a o f t h e n e ig h b o u r c o d e , w h ic h h a s s o f a r p r o v e d i t s e l f u s e f u l a s
a r e p r e s e n t a t i o n o f a se g m e n te d p i c t u r e .
I f o n e m ay m e rg e a n d a n a l y s e r e g i o n s o n a b a s i s o f t h e i r n e i g h b o u r s , c a n
we n o t th e n d e v e l o p a r e p r e s e n t a t i o n , l e s s b u lk y th a n t h a t o f t h e e n t i r e "
p i c t u r e , t h a t w i l l h o ld t h i s i n f o r m a t i o n , a n d i n w h ic h o n e may r e p r e s e n t
m e rg e s w i t h o u t r e v e r t i n g t o t h e o r i g i n a l p i c t u r e ?
c a n in d e e d d o s o .
The r e p r e s e n t a t i o n
is
I t t u r n s o u t t h a t we
fo rm e d s im p ly b y l i s t i n g
th e
n e ig h b o u r s i n t h e o r d e r i n w h ic h t h e y o c c u r i n a t r a v e r s a l o f t h e o u t s i d e
b o u n d a ry o f t h e r e g i o n , a n d i s c a l l e d t h e n e ig h b o u r c o d e .
NEIGHBOUR CODES.
PROPERTIES OF THE NEIGHBOUR CODE.
T he
f irs t
q u e s tio n t h a t o c c u rs i s ,
"Why o n ly t h e o u t s i d e n e ig h b o u r s ? "
T h e a n s w e r i s s im p l e : " i f a r e g i o n i s e n t i r e l y c o n t a i n e d b y a n o t h e r , t h e
in f o r m a tio n i s
p re s e rv e d in
need n o t b e e n te re d
in to
t h e c o n t a i n e d r e g i o n ' s n e ig h b o u r c o d e ,
an d
t h e o t h e r ' s c o d e , w h e re i t c o u ld c a u s e c o n s i d ­
e r a b l e b o o k - k e e p in g p r o b le m s a t some l a t e r
s ta g e ."
We i g n o r e i n t e r n a l
r e g i o n s w hen f o rm in g t h e c o d e ; t h a t d o e s n o t m ean t h a t t h e r e g i o n s o n t h e
in s i d e
r e c e iv e no c o n s i d e r a t i o n , n o r t h a t t h e i r r e l a t i o n s h i p w ith t h e i r
e n c l o s i n g n e ig h b o u r i s n o t r e c o r d e d , b u t t h a t s u c h c o n s i d e r a t i o n i s p e r ­
fo rm e d when we i n s p e c t t h e e n c l o s e d r e g i o n s ' o u t s i d e n e ig h b o u r s .
c o n s i s t e n t r e p r e s e n ta tio n o f th e r e g io n s p a c e , i t
For a
i s , h o w e v e r, important
t o t r a v e r s e e a c h r e g i o n ' s b o u n d a ry i n t h e sam e d i r e c t i o n ( i . e .
e ith e r a ll
c lo c k w is e o r a l l a n t i - c l o c k w i s e ) . T h e c o d e i s c y c l i c ; i t i s n o t i m p o r ta n t
to
s t a r t t h e t r a v e r s e w i t h a n y p a r t i c u l a r n e ig h b o u r .
p i c t u r e m u st b e r e g a r d e d a s a r e g i o n
th e p ic tu re ,
and i t s
The edge o f th e
t h a t s u r ro u n d s a l l th e r e g io n s in
o c c u r r e n c e i n t h e n e ig h b o u r c o d e m u st b e r e c o r d e d ;
i n t h i s w o rk , we h a v e a s s i g n e d t h e e d g e t h e r e g i o n n u m b er - 1 .
t h e ed g e i s im p o r ta n t,
a s we s h a l l s h o w .)
The code i s
o f t h e r e g i o n n u m b e rs o f n e ig h b o u r i n g r e g i o n s ,
fo rm e d o f t h e i r t y p e s , t e x t u r e s , e t c .
(R e c o r d in g
n o r m a l ly
fo rm ed
b u t may e q u a l l y w e l l b e
I t a p p e a rs t o b e p o s s ib l e t o c o n ­
s t r u c t an u n a m b ig u o u s r e p r e s e n t a t i o n o f t h e s e g m e n te d p i c t u r e w i t h t h i s
r e p r e s e n ta tio n , b u t I c an n o t p ro v e t h i s .
NEIGHBOUR CODES.
_
i.1
FORMING THE NEIGHBOUR CODE.
We a s s i g n e a c h c o n n e c te d r e g i o n i n t h e se g m e n te d p i c t u r e a u n iq u e s y m b o l,
c o n v e n t i o n a l l y a p o s i t i v e i n t e g e r , b y w h ic h we i d e n t i f y i t , w h ic h we u s e
h e r e , we r e f e r t o
th is
sy m b o l a s t h e r e g i o n n u m b e r.
The c o d e i s s t r u c ­
t u r e d a s f o l l o w s : a n e ig h b o u r t h a t a p p e a r s o n t h e b o u n d a ry o f .th e r e g i o n
in q u e s tio n
b e tw e e n i t
is
r e p re s e n te d once in
t h e n e ig h b o u r c o d e f o r e a c h c o n t a c t
an d t h e c e n t r a l r e g i o n , h o w ev er long th e b o u n d a ry on w hich
th e y c o n ta c t . T h u s t h e f o l l o w i n g a r e n o t l e g a l n e ig h b o u r c o d e s :
*
1 2 3 3 3 4 5 6
•
1 2 3 4 5 6 1
( r e c a l l i n g t h a t th e code i s c y c lic
•
).
T he f o llo w i n g c o d e s a r e ,h o w e v e r .
1 2 3 4 3 5 6
1 2 3 4 1 5 6 7
I t s h o u ld b e n o t e d t h a t a c o d e o f t h e fo rm :
8
•
is l e g a l ,
1 2 3 4 3 4 5
b u t p o te n tia lly c o n fu sin g ,
a n d n o t m e a n in g f u l ( s e e f i g u r e 2 ) .
T he i l l e g a l c o d e s o c c u r o c c a s i o n a l l y t h e m e rg in g p r o c e s s , an d r e q u i r e o n ly
a s im p le c l e a n i n g - u p p r o c e s s :
E
NEIGHBOUR CODES.
t h e a b o v e m e a n in g l e s s c o d e . m is r e p r e s e n t s
Codes of th e form:
I 2 3 4 3 4 5
a
<
.'"vl
th e s p a c c ,
a n d co n
a ri& c
c o d e s c a n n o t , h o w e v e r,
throi)f;l> i n c o r r e c t fo rrao L io n o f th o c o d c .
a r i s e w hen t h e c o d e i s
c o rr e c tly
Such
f o rm e d , o r t h e
r e p r e s e n t a t i o n i t s e l f w o u ld n o t b e u s e f u l .
We h a v e n o t s e e n b e f o r e t h e
c o n fu sio n t h a t can a r is e
o f w h e t h e r t h e se g m e n te d p i c t u r e i s
w ill i l l u s t r a t e th e p o in t.
as
e ig h t- c o n n e c te d ,
its
fro m t h e q u e s t i o n
fo u r- o r e ig h t- c o n n e c te d .
F ig u re 3
T h e p r o b le m i s s im p le ; i f we d e f i n e a r e g i o n
n e ig h b o u r s
m u st th e n b e f o u r - c o n n e c t e d .
T he
p r o b le m th e n a r i s e s t h a t i t i s n o t e a s y t o s a y i n c e r t a i n , c a s e s w h e th e r
a p i x e l i s an o u t s i d e n e ig h b o u r o r a s m a ll e n c lo s e d r e g i o n .
F o r tu n a te ly ,
we c a n s a y t h a t t h e i s s u e w i l l b e d e a l t w ith b y t h e l a b e l l e r , b y r e q u i r i n g
th a t i t
l a b e l c o n n e c te d r e g i o n s w i t h a s i n g l e l a b e l p e r r e g i o n ( a ll o w i n g
i t t o d e c i d e w h e t h e r a p i x e l b e lo n g s o n i t s ow n, o r w i t h a c o n n e c te d r e ­
g io n ) a n d b y th e n
in f o r m in g i t w h e t h e r i t s h o u ld d e f i n e e ig h t - c o n n e c t e d
r e g io n s o r f o u r- c o n n e c te d r e g io n s .
T h e l a t t e r d e c i s i o n we h a v e made a r ­
b i t r a r i l y : B e rg er
( 1 9 8 5 ) h a s b e e n a b l e t o p r o d u c e m e a n in g f u l
c o d es
se g m e n te d
fro m
th e
(SP ID E R ,1 9 8 4 ) , a n d t h e r e
p ic tu r e
a p p e a rs to
p ro d u ce d
by
a
n e ig h b o u r
package
r o u tin e
b e n o g o o d r e a s o n why h i s
ro u tin e s
s h o u ld n o t w o rk f o r f o u r - c o n n e c t e d r e g i o n s .
I t w i l l b e s e e n when f i g u r e 3 i s c o n s u l t e d t h a t i t
w h ic h d i r e c t i o n
e a c h b o u n d a ry p i x e l ' s
a r e f o rm in g t h e c o d e .
g iv e n ( 'e r e c t i o n :
i s a l s o im p o r t a n t i n
n e ig h b o u r s a r e t r a v e r s e d when we
We fo rm t h e c o d e b y t r a v e r s i n g t h e b o u n d a ry i n
i f we h a v e d e f i n e d
com p lem en t ( t h e r e s t o f t h e p i c t u r e )
th e re g io n as e ig h t- c o n n e c te d ,
a
its
i s t h e n f o u r - c o n n e c t e d , s o we m u st
th e n i n s p e c t e a c h b o u n d a ry p i x e l ' s f o u r n e ig h b o u r s .
T he t r a v e r s e o f t h e s e
f o u r n e ig h b o u r s m u st b e p e rf o rm e d i n t h e c o r r e c t d i r e c t i o n , s t a r t i n g
in
t h e r e g i o n w hose n e ig h b o u r c o d e we a r e 'f o r m i n g , a s f a i l i n g t o d o t h i s w i l l
l e a d t o s p u r io u s r e v e r s a l s i n t h e c o d e .
NEIGHBOUR CODES.
F o r e x am p le :
•
i n loimSr.g a c o d e , v c c a n jnc3u.dc> e a c h n c jg lib o u r p i x e l
and tlio n
c l e a n t h e c o d a u p , by c o m p re s s in g e a c h r u n c f r e g i o n n u m b e rs t o
a s i n g l e m em ber,
an d s u p p r e s s i n g t h e
r e g i o n num ber a t t h e end
o f t h e c o d e i f i t i s t h e same a s t h a t a t t h e b e g in n in g , s o t h a t
-
1 1 1 2 2
1 33
3 4 4 4 5 5 5 6 6 6 1 1 1
b eco m es
•
h o w e v e r, i f we h a v e m i s t r a v e r s e d p i x e l s , we may g e t
-
1 1 1 2 1 2 2 3 2 3 3 4 3 4 4 5 4 5 5 5 6 5 6 6 6
w h ic h i s u n m a n a g e a b le an d c e r t a i n l y e v e n w ith c l e a n i n g - u p c o u ld
n o t r e p r e s e n t a s e t o f c o n n e c te d r e g i o n s o n a p l a n e .
R e feren c e
t o f i g u r e s 2 a n d 3 s h o u ld rem o ve a n y c o n f u s i o n .
If,
h o w e v e r, we c a n n o t m e rg e two regions using t h e i r n e ig h b o u r c o d e s
a l o n e ' t o f i n d t h e c o d e f o r t h e c o m p o s ite r e g i o n , t h e r e p r e s e n t a t i o n
is
of little
th e
case,
u se.
F o r some tim e i t
u n til a s u ffic ie n tly
a p p e a r e d t h a t t h i s was in d e e d
p o w erfu l
a lg o r i t h m
w as
d e v e lo p e d ..
T h e m e rg in g o f c o d e s i s n o t co m p lex u n l e s s t h e tw o c a n d i d a t e r e g i o n s
i/ill
fo rm a c o m p o s ite t h a t e n c l o s e s
a n o th e r s e t o f r e g i o n s , w h ic h
w e re n o t p r e v i o u s l y e n c l o s e d , i n w h ic h c a s e t h e p r o b le m a p p e a r s t h a t
it
i s n o t e a s y t o t e l l w h ic h o f t h e g r o u p s r e p r e s e n t e d i n t h e c o d es
w i l l be e n c lo s e d .
.NEIGHBOUR CODES.
We s h a l l e x am in e t h e s im p le c a s e f i r s t .
%
%
r
A pixel's 4 " n e i g h b o u r s
A pixels S - n & l^ h b o u r s
A region -is n - c o n n z c t& d if one. m a y
ra o c h a m j one.
p o int in i t fro m a n y other; by a p a th
w h o se
next
of th e c u r r e n t m e m b e r.
m c rn b e r is ur, n - n e i a h b o u r
If we
t r a v e r s e pixels a s A . vnc will see.
I f w e trav e rse pixels a s B, w e wi)l s e e
If we
Fie.3
-travarsfc pixels o s
C, t h e
o ^ d th e n
3 a < ja m .
5 , 2 a n d th e n
5 , 4
5 a g a i 'n .
code, willbe c o r re c t.
The importance o f the direction
of pixel travarsal.
•
GJvcsi t l i e i we w ish t o m e rg e r e g i o n s 3 an d 2 w ith c o d e s
6 an d 7 8 1 9
c o m p o s ite
2 3
r e s p e c t i v e l y , an d t h a t 1 i s t h e s e n i o r r e g i o n
w ill
be
c a lle d
1,
an d
we a r e m e rg in g 2 t o
4 5
( th e
1) ,
bo
p r o d u c e t h e c o d e f o r t h e c o m p o s ite , We r o t a t e e a c h r e g i o n ' s c o d e
‘ s u c h t h a t t h e o t h e r m e rg e c a n d i d a t e a p p e a r s i n t h e f r o n t o f t h e
c o d e , s o t h a t we g e t
•
—
fo r 1 : 2 3 4 5 6
-
fo r 2 : 1 9 7 8
.
We th e n rem o v e fro m e a c h c o d e t h e
f i r s t m em ber,
and w r i te
one
c o d e a f t e r t h e o t h e r ( i t i s n o t im p o r t a n t w h ic h c o d e i s w r i t t e n
f irs t)
an d t h i s b eco m e s t h e c o d e f o r t h e c o m p o s ite . So we g e t :
3 4 5 6 9 7 6
We now c o n s i d e r t h e m o re i n t e r e s t i n g c a s e w h e re t h e m erg e c a n d i d a t e s
w ilt, e n c l o s e o n e o r m o re s e t s o f r e g i o n s w hen m e rg e d .
On r e f l e c t i o n ,
o r o n c o n s u l t a t i o n o f f i g u r e 4 , i t w i l l b e s e e n t h a t when t h e r e g io n s
w ill
ti m e s .
tim e s,
e n c lo s e
n s e t s o f r e g i o n s , th e y w i l l c o n t a c t e a c h o t h e r n +1
T h u s , e a c h c a n d i d a t e w i l l r e f e r t o t h e o t h e r i n i t s 'c o d e n+1
an d o n l y o n e o f t h e g r o u p s o f
r e g io n s
d e lim ite d
by
th e se
r e f e r e n c e s w i l l r e f e r t o r e g io n s t h a t w i l l b e on th e o u ts id e o f th e
c o m p o s ite r e g i o n :
th e o th e r gro u p s r e f e r to
e n c l o s e d a n d h e n c e s h o u ld n o t a p p e a r i n t h e
f e r e n t i a t e b e tw e e n t h e s e g r o u p s ?
NEIGHBOUR CODES.
re g io n s t h a t w i l l be
code.
How d o we d i f ­
It is
in f a c t f a i r l y
s im p le t o f i n d w h ic h g r o u p s o f r e g i o n s a r e e n ­
c l o s e d when we n o t i c e
e n c lo s e d in e ac h s e t .
t h a t o n ly
a f in ite
n u m b er o f r e g i o n s c a n b e
A lso i m p o r ta n t I s t h e f a c t c h a t we may 'c h a i n '
a c r o s s t h e r e p r e s e n t a t i o n t o f i n d a r e g i o n w h ic h h a s t h e e d g e o f t h e
p ic tu r e
a s a n e ig h b o u r,
s im p ly b y r e c o r d i n g
a ll
t h e n e ig h b o u r s o f
o u r g iv e n r e g i o n , a n d t h e i r n e ig h b o u r s , a n d s o o n , u n t i l we f i n d t h e
edge.
T he a l g o r i t h m t o m e rg e n e ig h b o u r c o d e s w i t h e n c lo s e d r e g i o n s
c a n th e n be e x p re s s e d a s ;
•
R o t a t e b o th c o d e s s o t h a t
t h e i r f i r s t member i s t h e o t h e r m erg e
c a n d id a te .
a g r o u p o f r e g i o n s b e tw e e n t h e f i r s t
We t h e n h a v e
an d t h e s e c o n d i n s t a n c e o f t h e c o d e s , w h ic h we r e c o r d s e p a r a t e l y
f o r each code.
•
Now we w is h t o d e c i d e f o r
e a c h c a n d i d a t e w h e th e r t h i s g r o u p r e ­
p r e s e n ts a n e n c lo s e d r e g io n o r n o t.
b e c o n s id e re d s e p a r a t e ly
N o tic e t h a t t h e g r o u p s m ust
f o r e a c h c a n d i d a t e a s we h a v e n o way
o f k n o w in g w h e t h e r t h e g r o u p s c o r r e s p o n d t o t h e s s v e s e t o f r e ­
g io n s .
We t h e n , f o r e a c h g r o u p , lo o k f o r t h e e d g e o f t h e p i c t u r e
i n t h e m a n n er d e s c r i b e d a b o v e : i f we f i n d i t , th e n t h e g r o u p m u st
r e p r e s e n t a n o u t s i d e g ro u p ;, i f we d o n o t , b u t r a t h e r e n u m e r a te
a f in ite
s e t o f r e g io n s
(w h e re
no
new
r e g io n s
a re
found
by
lo o k i n g a t new r e g i o n s ' n e ig h b o u r c o d e s , a n d a f t e r a w h ile t h e r e
a r e o n l y r e g i o n s w h o se n e ig h b o u r c o d e s h a v e b e e n in s p c J
L) th e n
we h a v e an i n s i d e g r o u p .
•
I f t h e g r o u p w as an i n s i d e g r o u p , we d i s c a r d j t b y r o t a t i n g t h e
co d e so t h a t th e seco n d
NEIGHBOUR CODES.
in s ta n c e o f th e o th e r c a n d id a te i s in
2J
.s id e group, wo di&card thi
i f t h e c o d e , and t h a t c o d e i s
r e a d y f o r m e rg in g .
r e a d y f o r m e rg in g , ( e a c h w i l l h a v e O nly
i n s t a n c e o f t h e o t h e r c a n d i d a t e 's r e g i o n n u m b e r) th e y
way
m erg ed
is- s im p le c o d e s .
F i g u r e s 4 a n d 5 may c l a r i f y t h i s e x p l a n a t i o n .
Our p i c t u r e h a s p r o d u c e d t h e
We g i v e a n e x am p le b e lo w .
f o llo w i n g n e ig h b o u r c o d e s :
(we
in d ic a te th e edge)
w is h t o m e rg e r e g i o n s 3
The f i r s t g r o u p f o r 4 i s 1 2
(3 i s n o t i n c lu d e d ; we a r e m e rg in g
s a l l g r o u p s w o u ld th e n b e
o u ts id e g r o u p s ) .
Now f o r L
n e ig h b o u r s - 1 3 an d 2 :
i s a c a n d i d a t e , an d i s i g n o r e d ) t t -1 i n d i c a t e s t h e
g ro u p i s 't h e
w a n t, a n d t h e c o d e f o r 4 beco m es:
(4
fig .4
"Iwo reg io n s c o n ta c tin g in
n+1 p oin ts enclosing ■n se ts
cf nzgions.
The
in s id e , g r o u p :
o b ta in in g • n e w
c o rito a rv rn
fis.5
o n lu
os
we
n e ig h b o u rs ,
5
Q nd
6.
Thz, n e ig h b o u r co d e
example.
m erg in g
T he f i r s t
g ro up f o r
3 i s 5 6 a n d n o ru-k" iv ig h b o u r.s arc: io u n d fro ir LhcbO
r e g i o n s ' c o d e s , ( i g n o r i n g 3 a n d 4 f o r t h e sam e r e a s o n s a s a b o v e ) s o t h a t
tK 'is g r o u p i s
ail e n c lo s e d g r o u p an d we may rem ove i t
fro m t h e c o d e and
g e t f o r 3:
Now we may m e rg e 4 a n d 3 c o n v e n t i o n a l l y .
T h is r e p r e s e n ta tio n
is
im p le m e n te d f a i r l y
s im p ly b y k e e p in g a t a b l e o f
t h e c o d e s , a n d a c r o s s - r e f e r e n c e t a b l e i n w h ic h we may s t o r e i n f o r m a ti o n
a s t o w h ic h r e g i o n h a s b e e n m e rg e d t o w h ic h .
CONCLUSION.
t h e p o w er o f t h e n e ig h b o u r c o d e r e p r e s e n t a t i o n
is
th e s im p lif ic a tio n
p r o v i d e s i h t h e repres»!>V .»^*on o f t h e s e g m e n te d p i c t u r e .
it
T h e c o d e s form
a r e p r e s e n t a t i o n i n w h ic i, f t I s s im p le t o m e rg e r e g i o n s , a n d w h ic h i s i n
g e n eral b e tte r - s u ite d
to
t h e n e e d s o f a k n o w le d g e - in te n s iv e s y s te m th a n
i s t h e c o n v e n t i o n a l r e p r e s e n t a t i o n o f t h e se g m e n te d p i c t u r e .
NEIGHBOUR CODES.
.
31
THE PARADIGM AND ITS USE.
In th is
a u t h o r 's o p in io n ,
liv i n g
sy s te m s ,
and i s
v is io n
is
a f u n c t i o n v e r y d e e p ly e m bedded i n
n o t c o rre c tly
d e s c rib e d
a s an i n t e l l i g e n t
T h e p r o b le m o f g e n e r a l s c e n e a n a l y s i s d o e s h o w e v e r seem t o
re q u ire
a c t.
fo r
i t s s o l u t i o n t h e e m p lo y m e n t o f some fo rm o f k n o w le d g e , w h ic h a p p e a r s i n ­
t r i n s i c t o m o st s e e in g I tf e - f o r m s .
A lth o u g h we c la im t h a t k n o w le d g e w i l l
h a v e t o b e e m p lo y e d t o p r o d u c e a g e n e r a l s c e n e - a n a l y s i s s y s te m , we c a n n o t
c la im t h a t t h i s k n o w le d g e w i l l b e e a s y t o
fin d , o r to e x p re ss.
I t seem s
l i k e l y t o o t h a t t h e K S 's w i l l p r o l i f e r a t e , a n d t h e r e se em s t o b e n o g o o d
r e a s o n f o r e x p e c t i n g t h a t t h e k n o w le d g e b a s e w i l l n o t h o ld i n t e r n a l c o n ­
tra d ic tio n s .
W ith t h i s
i n m in d , we s t i l l
f s e l i t p r o d u c tiv e t o p ro p o se
a new p a r a d ig m , t h e m o t i v a t i o n f o r w h ic h h a s b e e n g iv e n a b o v e ,
i n w h ic h
s c e n e s may b e a n a l y s e d i n a k n o w l e d g e - i n t e n s i v e f a s h i o n .
GUIDING CO N C EPTS: THE PARADIGM ITSELF.
T he m a jo r t e n e t ,
a n d t h e th e m e o f t h i s d i s s e r t a t i o n , i s t h e b e l i e f t h a t
p i c t u r e s s h o u ld b e a n a ly s e d i n te rm s o f e x p l i c i t k n o w le d g e a b o u t p i c t u r e s ,
an d a b o u t w h at o b j e c t s
lo o k
m o d e ls made o f t h e p i c t u r e .
l i k e on p i c t u r e s , r a t h e r th a n
i n te r m s
of
A m o d el o f th e . w o r l d m u st b e u s e d , t o h a v e
s u f f i c i e n t k n o w le d g e t o i n t e r p r e t t h e p i c t u r e ,
b u t th e
p ic tu r e
need n o t
b e m o d e lle d , s c t h a t p i c t u r e - l e v e l o p e r a t i o n s may b e g u id e d b y k n o w le d g e
b o th a b o u t t h e p i c t u r e a n d a b o u t t h e w o r l d .
THE PARADIGM AND ITS USE.
T he se c o n d t e n e t i s t h a t t h e p i c t u r e m u st b e h a n d le d i n te rm s o f r e g i o n s :
th is
is
im p o r ta n t,
as
g ro w in g d o n o t e a s i l y
th e p ro c e s s e s
o f ed g e d e te c tio n and o f o u t l i n e
a llo w b a c k t r a c k i n g t o i n c o r p o r a t e w o rld k n o w le d g e .
T h is t e n e t a ssu m e s t h a t a s e g m e n t e r i s a v a i l a b l e , t h a t i s a b l e t o se g m e n t
th e p ic tu r e such t h a t e ac h d e fin e d p r im it iv e
o ne re g io n .
is
r e p r e s e n te d
by a t
le a s t
We e n v i s a g e b a c k t r a c k i n g t o t h e l e v e l o f u n d o in g i n c o r r e c t
m e rg e s ; we d o n o t a t p r e s e n t s e e b a c k t r a c k i n g t o
t a t i o n o f th e p i c t u r e
a le v e l o f a rese g m e n ­
( a l t h o u g h t h i s c o u ld b e a c h i e v e d , m ore s im p ly t h a n
b y t h e p r o c e s s w h ic h w o u ld a p p ly i f t h e p i c t u r e h a d b e e n i n t e r p r e t e d i n
te rm s o f i t s e d g e s ) .
T he em ploym ent o f k n o w le d g e s h o u ld o c c u r a t m any l e v e l s ,
fro m p l a n n i n g
t h e p r o c e s s b y w h ic h p r i m i t i v e s a r e m e rg e d , t o u s i n g t h e c o n t e x t i n w h ic h
t h e p i c t u r e i s p r e s e n t e d t o r e d u c e t h e q u a n t i t y o f k n o w le d g e r e q u i r e d t o
in te r p re t i t .
How th e n d o we i n t e n d a s y s te m t o f i n d a g iv e n o b j e c t i n a s c e n e ?
it
m u st
red u c e
th e
d e s c r ip tio n
of
th e
o b je c t
u n d e r s t a n d s , u s i n g t h e k n o w le d g e a v a i l a b l e t o i t .
I t m u st t h e n f i n d t h e s e
p r i m i t i v e s , w i t h t h e r e l a t i o n s h i p s b e tw e e n p r i m i t i v e s
If
it
F irs tly ,
‘o p r i m i t i v e s w h i- , i t
th a t i t
r e q u ir e s .
c a n n o t f i n d a l l o f t h e r e q u i r e d p r i m i t i v e s , i t s h o u ld i n s p e c t , t h e
s e t w h ic h i t
h as,
a n d d e c i d e w h e th e r i t
can s t i l l re a so n a b ly f in d
o b j e c t b y l e n i e n t r e c l a s s i f i c a t i o n o f som e r e g i o n s .
th e
T h u s , t h e s y s te m h a s
a d e f in e d g o a l s t r u c t u r e , w h ic h w i l l b e a s f o l l o w s :
•
F in d o u t w h a t s h o u ld b e f o u n d i n t h e p i c t u r e ( i . e . f i n d t h e m a in
sy s te m g o a l ) .
THE PARADIGM AND ITS USE.
•
F a c to r t h i s g o al i n to a l i s t o f p r im it iv e s and a d e s c r ip tio n o f
s tru c tu re ,
u s in g
a v a i l a b l e k n o w le d g e .
t h e s y s te m i s e n t i t l e d t o f a i l .
If
th is
is
im p o s s ib le ,
(How -w ould a human r e s p o n d when
a s k e d t o f i n d a n o b j e c t i n a s c e n e , w hen h e d o e s n o t know w h at
t h a t o b j e c t lo o k s l i k e , a n d t h e r e a r e m o re t h a n o n e unknow n o b ­
j e c t s in th e sc en e ?)
•
Now f i n d a s m any o f t h e s e p r i m i t i v e s i n t h e s c e n e a s p o s s i b l e ,
b y u s i n g k n o w le d g e t o g e n e r a t e a n d t e s t
m e rg e s o f p o s s i b l e c a n ­
d id a te r e g io n s , and t o c l a s s i f y th e s e r e g io n s o r g ro u p s o f r e -
•
F in a lly ^ , f in d a g ro u p o f p r im it iv e s th e
resp o n d s to
th e s tr u c tu r a l d e s c r ip tio n
s t r u c t u r e o f w h ic h c o r ­
s to r e d ;
if
a p r o m is in g
g r o u p i s m i s s i n g a p r i r . V t i v e , a t te m p t t o f i n d i t b y r e c l a s s i f y i n g
r e g i o n s w h e re t h e
p r im itiv e
is
s t r u c tu r e o f th e o b je c t
lik e ly
to o c cu r.
in d ic a te s
th a t
th e
I f t h i s g r o u p c a n b e f o u n d , th e
s y s te m g o a l h a s b e e n f o u n d , a n d t h e n e x t g o a l w ay b e r e q u e s t e d .
We h a v e s e e n t h a t t h e u s e o f r e g i o n s s h o u ld
b a c k tr a c k in g i s
s im p lif ie d ,
and th e
b e e n p r o d u c e d u n d e r t h e new p a r a d ig m .
a s y s te m ?
THE PARADIGM AND ITS USE.
s im p lify
th e
p ro ce ss,
as
s y s te m u s e s k n o w le d g e ; t h u s , i t h a s
B u t w h a t i s t h e s t r u c t u r e o f su c h
A SYSTEM DESIGNED TO ANALYSE SCENES
T he t a s k o f t h i s s y s te m w i l l b e t o f i n d a g iv e n o b j e c t
t o say t h a t th e o b je c t i s
n o t in
in
a p ic tu re , or
t h e p i c t u r e , b y e x a m in in g r e g i o n s a n d
g ro u p s o f r e g io n s w ith th e in t e n t i o n o f f in d in g a g ro u p t h a t b e s t c o r r e ­
sp o n d s t o t h e o b j e c t r e q u i r e d .
s y s te m g o a l . )
(We w i l l r e f e r t o t h i s o b j e c t a s t h e m ain
T h e s y s te m s h o u ld b e a b l e t o e v a l u a t e i t s a t t e m p t s t o m erge
r e g i o n s t o fo rm p r i m i t i v e s , a n d t o p l a n a b e s t c o u r s e o f a c t i o n b y w h ic h
i t . s h o u l d b e a b l e t o f i n d t h e m a in s y s te m g o a l .
To a c h ie v
t h e s e a im s ,
t h e s y s te m w i l l u s e k n o w le d g e ,
T he
s y s te m w i l l ,
th e n , have a
'l e d g e b a s e , w h ic h w i l l n e e d t o
c o n t a i n b o th w o r l d k n o w le d g e e n d p i c t u r e - l e v e l " k n o w le d g e .
T h e s y s te m w i l l
c o n s i s t o f s e c t i o n s , w h ic h w i l l b e s e l f - c o n t a i n e d , e n d w h ic h we s h a l l c a l l
K n o w le d g e -A p p lic a tio n B lo c k s o r K A B 's.
d i f f e r e n t f o n a o f k n o w le d g e ,
t h e e n t i r e k n o w le d g e b a s e .
E ac h KAB w i l l r e f e r t o a n d u s e a
and w i l l n o t in g e n e r a l r e q u ir e
acc ess to
H ow ever t h e KAB w h ic h u s e s w o r l d k n o w le d g e
t o f a c t o r t h e m a in s y s te m g o a l may n e e d a t some s t a g e t o h a v e a i o r c o f
c o n t r o l e x e r c is e d on i t s
k n o w le d g e b a s e , t o p r e v e n t i t
And we h a v e d i s c u s s e d t h e a v a i l a b l e s t r a t e g i e s
w o u ld a p p e a r t h a t t h e b e s t s t r a t e g y
c o r d in g t o i t s
is
above;
fro m s a t u r a t i n g .
in t h i s
case,
it
t o o r d e r t h e k n o w le d g e b a s e a c ­
r e l e v a n c e t o t h e m a in s y s te m g o a l .
T h u s, in a s e a rc h f o r
a t e l e p h o n e , k n o w le d g e a b o u t h a n d s e t s w i l l b e made m o re a c c e s s i b l e t h a n . '
kn o w le d g e a b o u t h a n d s .
We r e f e r t o t h e k n o w le d g e c o n v e y e d b y t h i s ty p e
o f s t r a t e g y a s " c o n t e x t u a l k n o w le d g e " , a s i t i s k n o w le d g e a b o u t t h e c o n ­
t e x t i n w h ic h t h e p i c t u r e a p p e a r s .
As we h a v e s a i d ,
e a s y t o im p le m e n t t h i s fo rm o f k n o w le d g e m an ag e m en t.
t h u s y s te m i s show n i n f i g u r e 6 .
THE PARADIGM AND ITS USE.
it
d oes n o t appear
A b lo c k d ia g ra m o f
K n o w le d g e - a p p lic a t io n b l o c k s , w i t h t h e i r
Fis, 6
A system designed undar the new
paradigm .
r e l e v a n t k n o w le d g e b a s e s , a r e shown a s c i r c l e s , a n d t h e flo w o f in f o r m a ­
t i o n i s shown b y a r r o w s .
KAB
co n tro l
T h e d u t i e s o f e a c h KAB w i l l b e d e s c r i b e d b e lo w .
DUTIES
T h is
b lo c k w i l l recom m end c h a n g e s o f c o n t e x t a n d
w i l l m anage t h e p r o c e s s b y w h ic h c o n t r o l i s p a s s e d
fro m KAB t o KAB.
c o n te x t se le c t
T h is b l o c k w i l l p r o v i d e c o n t e x t u a l i n f o r m a t i o n b y
o r d e r i n g t h e k n o w le d g e b a s e .
goal s t r u c t u r e s e tu p
T h is
b lo c k w i l l p u t to g e th e r a g o a l s t r u c t u r e t o
a llo w t h e f i n d i n g o f c o m p le x s t r u c t u r e s b y f i n d i n g
t h e i r c o m p o n e n ts , u s i n g a v a i l a b l e w o rld k n o w le d g e ,
an d w i l l
in f o r m t h e a c t i o n
e v a lu a to r a s to
s t r u c t u r e s s h o u ld b e f o u n d w h e r e ,
w h ic h
i f n e c e ssa ry by
le n ie n t c la s s ific a tio n .
actio n e v a lu a te
T h i s b lo c k w i l l e v a l u a t e
te r m s
any
m erge
s u g g e s te d
in
o f t h e p r o p e r t i e s o f t h e m erg ed r e g i o n a n d
o f w h a t i s b e in g s o u g h t.
ac tio n p a th e v a lu a te
T h is b lo c k w i l l e v a l u a t e a s e r i e s o f m e rg in g d e c i ­
s io n s
in
te rm s
k n o w le d g e ,
of
p ic tu re
k n o w le d g e
and
p o s e d m e rg e i n te rm s o f t h e o t h e r m e rg e s
fro m
THE PARADIGM AND ITS USE.
it.
w o rld
a n d w i l l recom m end o r p e n a l i z e a p r o ­
. It
a ris in g
s h o u ld b e a b l e t o d e c id e b e tw e e n tw o
5kV
p o s s i b l e m e rg e d e c i s i o n s i n te rm s o f t h e i r o v e r a l l
e ffe c t,
an d th e n r e p o r t t o t h e p a t h p l a n n e r .
G iv e n a r e g i o n w ith g i v e n p h y s i c a l p r o p e r t i e s , t h i s
b lo c k w i l l d e c id e w h at i t
T h is b lo c k w i l l a l s o
is
lik e ly
a tte m p t t o
to
r e p re s e n t.
la b e l s e ts o f r e ­
g i o n s , g iv e n a p p e a r a n c e a n d s t r u c t u r a l in f o r m a t i o n .
T h i s b lo c k w i l l p r o v i d e a m e rg in g p l a n t o m e e t t h e
g o al s tru c tu re ,
and w i l l o f f e r
a lte r n a tiv e
p a th s t o th e a c tio n p a th e v a lu a to r .
m erg e
T hus, i t w ill
i n t e r a c t w ith th e a c tio n p a th e v a lu a to r t o p ro v id e
an o v e r a ll s t r a t e g y .
T h i s b lo c k w i l l t a k e t h s a c t i o n s recom m ended b y t h e
p a th e v a lu a to r .
A s y s te m s t r u c t u r e d
a s d e s c r i b e d a b o v e s h o u ld b e a b l e t o a n a l y s e f a i r l y
g e n e r a l s c e n e s ; h o w e v e r,
it
w ill
c o n s t r u c t , a n d w o u ld a p p e a r t o
n iq u e s d e s c r i b e d
be
>s
%
s o p h is tic a te d
th e p r e s e n t l i t e r a t u r e .
To i l l u s t r a t e
f u n c tio n in g
sc en e .
•
of
th e
th e
in te n tio n
of th is
sys'.rftn w hen r e q u i r e d
te c h n i q u e s
to
c a p a b i l i t i e s o f th e te c h ­
U s e f u l in f o r m a t i o n
o b t a i n e d b y p r o t o t y p i n g a s y s te m w i t h som e o f t h e
s y s te m .
th e
in
re q u ir e
b e bey o n d th e
s h o u ld
fe a tu re s o f th is
d e s i g n , wo s h a l l d e s c r i b e
to
f in d
a te le p h o n e
in a
We a ssu m e a b l a c k d i a l t e l e p h o n e f o r a r g u m e n t's s a k e .
T he g o a l s t r u c t u r e se tu p 'K A B P i l l r e c e i v e t h e m a in s y s te m g o a l ,
an d f a c t o r
it,
f o r a r g u m e n t’s s a k e ,
" h a n d s e t " , a n d ’’h a n d s e t c o r d " .
THE PARADIGM A N D ‘ITS USE.
in to
" b o d y " , '"b o d y
"B o d y " w i l l b e f a c t o r e d
c o rd " ,
a g a in
i n t o " d i a l " an d "m ain b o d y " ; t h e o t h e r s u b - g o a ] s w i l l b e r e g a r d e d
as p rim itiv e s .
Thf, p r i m i t i v e s
a re d e s c rib e d a s
f o llo w s ,
fo r
s im p lic ity :
PRIMITIVE
DESCRIPTION
b od y co rd
d a rk ,
lo n g a n d t h i n ,
s m a ll,- lew r e f l e c t a n c e ,
low
te x tu re ,
h an d set
d a rk ,
b a n a n a -sh a p e d ,
m e d iu m -siz e d
,
medium
r e f l e c t a n c e , low t e x t u r e .
h a n d se t c o rd
d a r k , lo n g a n d c o i l e d , s m a l l , medium r e f l e c t a n c e ,
m edium t e x t u r e .
d ia l
light,
round
with
holes,
m e d iu m - s iz e d ,
medium
r e f l e c t a n c e , low t e x t u r e .
m a in b o d y
The p i c t u r e
d a r k , s q u a r e , l a r g e , low r e f l e c t a n c e , low t e x t u r e .
is
th e n s e g m e n te d , an d t h e c o n t r o l KAB w i l l
pass
c o n t r o l t o t h e p a t h p l a n n e r KAB, w h ic h w i l l p r o v i d e a p a t h o f .
( 'i i i s t y p e :
f i n d a n y r e g i o n s w h ic h a r e p r i m i t i v e s i n th e m s e lv e s ;
t h e n g e n a v s te m e rg e s b e tw e e n d a r k n e ig h b o u r i n g r e g i o n s .
T est
t h e s e m e rg e s w i t h t h e a c t i o n p a t h e v a l u a t e KAB, w h ic h w i l l te s 'c
in d iv i d u a l c a s e s w ith th e
a c tio n
e v a l u a t e KAB.
i s t o m e rg e to w a rd s a t l e a s t o d e p r i m i t i v e .
THE PARADIGM AND ITS U M ,
T he
in te n tio n -
A ssu m in g t h a t a p r i m i t i v e h a s b e e n f o u n d , so y th e h a n d s e t c o r d ,
t h e p a t h p l a n n e r KAB w i l l p a s s t h e a c t i o n p a t h e v a l u a t e KAB t h i s
i n f o r m a t i o n , b y in f o r m in g i t t h a t t h e h a n d s e t i t s e l f
is
lik e ly
t o b e f o u n d a s a n e ig h b o u r , o r a s a g r o u p c o n t a i n i n g a n e ig h b o u r ,
of th is
p r im itiv e .
S im ila r in f o r m a tio n w i l l be p a s s e d f o r th e
"m a in b o d y " p r i m i t i v e .
A ssume t h a t w i t h t h i s
new in f o r m a t i o n ,
t h e " m a in b o d y " p r i m i t i v e i s now f o u n d .
T h e p a t h p l a n KAB now may p a s s t o t h e a c t i o n p a t h e v a l u a t e b lo c k
a new s t r a t e g y , w h e re t h e l i k e l y p o s i t i o n o f t h e d i a l i s a ssu m e d ,
an d t h e l i k e l y p o s i t i o n o f t h e c o r d i s o f f e r e d .
b o d y c o r d i s now f o u n d : t h e h a n d s e t i s s t i l l
A ssume t h a t t h e
la c k in g .
T h e c o n t r o l KAB w i l l now p a s s c o n t r o l t o t h e g o a l s t r u c t u r e s e t u p
KAB, w h ic h w i l l i n s p e c t i t s w o rld k n o w le d g e a n d in f o r m t h e p a t h
p l a n n e r KAB, th r o u g h t h e c o n t r o l KAB, t h a t t h e h a n d s e t may b e
e i t h e r o n t h e m a in b o d y , a t t h e o t h e r e n d o f t h e h a n d s e t c o r d ,
o r a b s e n t, w ith lik e lih o o d s f o r e a c h s i t u a t i o n .
F in a lly ,
t h e p a t h p l a n n e r , h a v in g h a d a m erg e p a t h a c c e p te d b y
t h e a c t i o n p a t h e v a l u a t e KAB, a n d w i t h t h e l a b e l e tte m p '; KAB n o t
h a v in g f o u n d t h e h a n d s e t u s i n g t h i s
p a th , w ill p a ss th e e n tir e
s e t o f r e g i o n s t o t h e l a b e l a t te m p t KAB, a s i t know s t h a t i t n e e d
not
f in d a h a n d s e t.
T h e l a b e l a t t e m p t KAB w i l l t h e n u s e know ­
le d g e l e f t t o i t b y t h e g o a l s t r u c t u r e s e t u p b lo c k KAB t o l a b e l
t h e s e t o f r e g i o n s (w h ic h a r e a l s o p r i m i t i v e s : we m e rg e r e g io n s
in t h e r e p r e s e n t a t i o n i n t h e p r o c e s s o f m e rg in g u p t o t h e p r im ­
itiv e
fin d
le v e l,
a n d o n ly
s tr u c tu re s )
THE PARADIGM AND ITS (JSE.
as
i n a s c r a t c h p a d fo rm w hen we a t te m p t t o
a
te le p h o n e .
* I f th is
la b e llin g
a tte m p t
fa ils ,
th » p ro c e ss s t a r t s
a g a i n , w ith
a d i f f e r e n t s e t o f d a rk
re g io n s b e in g t r i e d a s a s t a r t to th e p r o c e s s .
I t c a n b e s e e n t h a t t o a l a r g e e x t e n t t h e c o n t r o l KAB a c t s a s a s c h e d u l e r :
in
fa c t,
in
th e
p r o to ty p e ,
th e
c o n tro l
KAB h a s
b e e n im p le m e n te d b y
s t r u c t u r i n g t h e w ay t h a t t h e KAB's a r e in v o k e d , r a t h e r th a n b y c o n f i g u r i n g
e s e p a r a t e KAB.
T he s y s te m a s d e s c r i b e d s h o u ld b e a b l e t o a n a l y s e s c e n e s w ith c o n s i d e r a b l e
g e n e r a l i t y , i f i t c a n b e im p le m e n te d e f f e c t i v e l y .
C e rta in ly , th e s tr u c ­
t u r e o f tbs» s y s te m i s s u c h t h a t i f i t i s a b l e t o a n a l y s e a s c e n e , i t w i l l
b e a b le t o a n a ly s e a t l e a s t a c la s s o f s c e n e s .
T h e r e a r e , h o w e v e r, p o ­
t e n t i a l p r o b le m s w i t h t h i s d e s i g n .
P O T E N T I A L PROBLEMS
It
m u st b e r e a l i s e d a t o n c e t h a t t h e d e s i g n d e s c r i b e d i s
p a p e r d e s ig n .
a p e n c i l an d
I t i s t h e p r o d u c t o f v e r y r e c e n t th o u g h t , an d h a s n o t b e e n '
p r o v e n , im p le m e n te d , o r d e e p ly t e s t e d , a n d a s s u c h s h o u ld b e e x p e c te d t o
have e r r o r s in i t .
A v e r y s im p le v e r s i o n o f t h i s sy s te m h a s b e e n p r o t o ­
t y p e d , b u t , a lth o u g h i t show s p r o m iu in g c h a r a c t e r i s t i c s , i t d o e s n o t p r o v e
th e v a l i d i t y o f th e d e s ig n .
f e e ls i t
sc en e a n a ly s i s ;
re s u lts .
V is io n i s a co m p lex a c t i v i t y , a n d t h i s a u th o r
u n l i k e l y t h a t t h e d e s ig n p r e s e n t e d h a s s o lv e d t h e
it
is
p r e s e n te d
p r o b le m
of
a s a new a p p r o a c h , w h ic h p r o m is e s good
I am c o n f i d e n t t h a t t h e t e n e t s o f t h e p a ra d ig m p r e s e n t e d a r e
s o u n d , h o w e v e r, a n d t h a t t h e y a r e p o t e n t i a l l y p r o d u c t i v e .
THE PARADIGM AND ITS USE.
40
T h e k n o w le d g e b a s e i s
lik e ly
to
i t s g ro w th may b e c o n t r o l l e d .
s y s te m s i s
be huge,
and i t
i s n o t e a s y t o s e e how
T he k n o w le d g e w h ic h d r i v e s human v i s i o n
n o t k n o w n , an d i t may b e s o d e e p ly 'e m b e d d ed t h a t t o d i s c o v e r
i t a t a l l w i l l r e q u i r e a r o u n d a b o u t a p p r o a c h i n v o l v i n g an o p e r a t o r g r a d in g
t h e r e s p o n s e o f t h e s y s te m t o s c e n e s .
I t i s n o t a fo rg o n e c o n c lu s io n t h a t
t h i s k n o w le d g e c a n b e d i s c o v e r e d a t a l l , a lth o u g h i t d o e s n o t a p p e a r im ­
p o s s ib le .
T h e r e s u l t s o f t h e l^ . o t o t y p i n g e x e r c i s e a r e d e s c r i b e d b e lo w , a n d i t w i l l
be
sh o w n
th e re
th a t
hope
e x is ts
fo r
th e
p r o d u c tio n
of
a
.g e n e r a l
s c e n e - a n a l y s i s s y s te m .
CONCLUSION
A p a ra d ig m f o r t h e a n a l y s i s o f g e n e r a l s c e n e s h a s b e e n d e s c r i b e d , a n d we
h a v e show n t h e d e s i g n o f a s y s te m t h a t w i l l p e r f o r m t h i s t a s k u n d e r t h i s
p a ra d ig m .
C r i t i c i s m s o f t h e p a ra d ig m a n d o f t h e d e s ig n h a v e b e e n p r e ­
s e n te d , b u t i t
is
f e l t th a t fu rth e r
w o r t h w h i le g o a l .
THE PARADIGM AND ITS U S E .
d e v e lo p m e n t o f
th is
s y s te m
'i s
a
THE DESIGN OF THE PROTOTYPE SYSTEM.
T h e s y s te m d e s c r i b e d
above h a s b e en p ro to ty p e d
in
a v e r y l i m i t e d fo rm ,
i n a n a tte m p t t o d e m o n s tr a te t h e v a l i d i t y o f t h e p a ra d ig m an d t o show t h a t
u s e f u l s y s te m s may b e d e s i g n e d w i t h i n
s tr u c t e d w ith e v e ry s i m p l i f i c a t i o n
it.
t h e r e a d e r w i l l h a v e g a t h e r e d fro m t h e
is s u e s have been s id e s te p p e d .
scen es, is
a n d t h e n e ig h b o u r c o d e
im p le m e n te d i n
t h i s p r o to ty p e h e r e , and i t s
t o a llo w i t s
c o n s tru c tio n ;
in tr o d u c tio n t h a t
many
co m p lex
T h e p r o t o t y p e , a lth o u g h n o t y e t a n a l y s i n g
s h o w in g some p r o m is e ,
has been e ff e c tiv e ly
T he p r o t o t y p e h a s b e e n c o n ­
p o s s ib le ,
it.
r e p re s e n ta tio n
We w i l l d i s c u s s t h e d e s ig n o f
im p l e m e n ta tio n a n d t h e r e s u l t s t h a t s u p p o r t
t h e v a l i d i t y o f t h e p a ra d ig m i n a l a t e r c h a p t e r .
T h e p r o t o t y p e w as i n t e n d e d , n o t t o d i s p l a y t h e m ore a d v a n c e d o f t h e c o n ­
c e p ts
d is c u s s e d
above,
nor
to
t e s t t h e w h o le o f t h e s t r u c t u r e o f t h e
s y s te m d e s c r i b e d a b o v e , b u t t o d e m o n s tr a te t h e d e s i r a b i l i t y o f c o n t i n u i n g
w ith w o rk o f t h i s
p a ra d ig m
w ith
s c rib e d above.
th e
ty p e i n
th is
in te n tio n
fie ld
and,
in
th is
o f t h e d e s ig n d e ­
As a r e s u l t , n o t. a l l t h e KAB's w e r e im p le m e n te d , an d th o s e
t h a t w e r e , w e re im p le m e n te d i n a l i m i t e d
w ere:
m o re p a r t i c u l a r l y ,
o f s h o w in g t h e v a l i d i t y
g o a l s tr u c tu re s e tu p
( c a lle d
fo rm .
T h e KAB's
im p le m e n te d
f g o a l ) , a c tio n p a th e v a lu a te , a c tio n
e v a l u a t e , a c t o r a n d a fo rm o f l a b e l a t te m p t ( g t m e r ) .
T h , 'd e s i g n o f t h e
p r o t o t y p e w i l l b e d i s c u s s e d u n d e r f o u r h e a d in g s : t h e b lo c k d e s i g n o f t h e
s y s te m , t h e d e s i g n o f t h e
d a t a s t r u c t u r e s , t h e d e s ig n o f t h e r u l e s , an d
t h e a l g o r i t h m s u s e d i n t h e K A B 's.
THE DESIGN OF THE PROTOTYPE SYSTEM.
P ictu re dcdo.
fis . 7
Block diagram of
system.
the prototype
THE BLOCK DESIGN OF THE PROTOTYPE SYSTEM.
We h a v e s i m p l i f i e d t h e c o n s t r u c t i o n o f t h e s y s te m d e s c r i b e d a b o v e b y d i ­
v id in g th e d u tie s
r e s p e c tiv e ly :
o f t h e KAB's i n t o tw o c l e a r b l o c k s , w h o se d u t i e s a r e
f i n d t h e d e s c r i p t i o n o f t h e m ain s y s te m g o a l i n te rm s o f
p r im it iv e s , ta d fin d th e g o a l d e s c rib e d in th e p ic tu r e .
we c a l l f g o a l , t h e s e c o n d g tm e r .
T h e f i r s t b lo c k
We w i l l d e s c r i b e h e r e t h e d u t i e s an d
t h e t e c h n i q u e s u s e d i n t h e s e b l o c k s , a s w e l l a s t h e tw o o t h e r b lo c k s t h e
s y s te m
has
in h e rite d
fro m
its
i m p l e m e n ta tio n :
s ta r t,
in its ,
a nd th e
lo a d e r , m r in t.
R e f e r t o f i g u r e 7 , w h ic h show s t h e b lo c k d ia g ra m o f t h e p r o t o t y p e s y s te m ;
t h e im p l e m e n ta tio n h a s r e q u i r e d t h e i n t r o d u c t i o n o f a f u r t h e r b l o c k , n o t
show n o n t h e f i g u r e , f o r i n i t i a l i s i n g d a t a s t r u c t u r e s .
N o te t h a t we do
n o t h a v e h e r e a s i t u a t i o n w h e re e a c h b l o c k a c t s o n d a t a , a n d t h e n p a s s e s
t h i s d a t a t o a f u r t h e r b lo c k ( h e n c e t h e fo rm o f t h e b lo c k d i a g r a m ) .
s te a d , th e
l o a d e r , known a s m r i n t
( t h e p ro g ra m n a m e ),
In ­
l o a d s e a c h b lo c k
a s a s e g m e n t u n t i l t h e b lo c k t e r m i n a t e s , in f o r m in g t h e lo a d e r w h ic h b lo c k
i s to be used n e x t.
E ac h b l o c k m u st t h e n t e r m i n a t e , h a v in g c h a n g e d t h e
sy s te m v a r i a b l e w h ic h in f o r m s m r i n t w h ic h b lo c k s h o u ld b e lo a d e d n e x t (we
c a ll th is
t h e s y s te m b l o c k c o n t e x t , " o r SB C ).
T h us t h e o p e r a t i o n o f t h e
s y s te m i s a s f o l l o w s :
•
To s t a r t t h e s y s te m , m r i n t i s in v o k e d .
d is k ,
w ith o u t
c o n s u ltin g
In tu r n m rin t r e t r i e v e s
t h e SBC, t h e p i c t u r e
re p re s e n ta tio n
from
le ft
t h e r e b y t h e r e p r e s e n t e r p r o g ra m , a n d th e n lo a d s t h e i n i t s b l c k ( n o t
show n i n
fig u re
7 ) , w h ic h i n i t i a l i s e s
th e
s y s te m d a t a s t r u c t u r e s .
‘ T h is b l o c k w i l l a lw a y s t e r m i n a t e " h a p p y " , u n l e s s t h e r e a r e h a rd w a r e
THE DESIGN OF THE PROTOTYPE SYSTEM.
e r r o r s , an d w i l l n o t f l f f e c l t h e
SBC.
T h e SBC is th e n s e t t o
s ta rt
by m r in t.
•
M r in t
th e n
r e tr ie v e s
th e
firs t
KS fro m d i s k .
E ach KS
f i e l d , a n d i f t h i s f i e l d d o e s n o t a g r e e w ith t h e c u r r e n t
i s d i s c a r d e d , a n d t h e n e x t KS r e t r i e v e d .
h a s a n SBC
SBC, t h e KS
T he SBC a t t h i s s t a g e w i l l
b e " s t o r k " , avid w h e n .a KS i s f o u n d w ith t h i s v a lu e i n i t s
t h e s t a r t b l o c k i s in v o k e d , a n d t h e KS p a s s e d t o i t .
h a s f i n i s h e d w i t h t h e K S, i t r e t u r n s c o n t r o l t o
SBC f i e l d ,
When t h i s b lo c k
m r in t.
T he p ro c e s s
c o n t i n u e s u n t i l t h e s t a r t b l o c k c h a n g e s t h e SBC, w h ic h i t w i l l a lw a y s
change to " fg o a l" .
•
T h e SBC i s
now " f g o a l " ; m r i n t w i l l c o n t i n u e t o in v o k e K S 's , d i s c a r d
th e m i f t h e i r SBC f i e l d d o e s n o t a g r e e w i t h t h e c u r r e n t SBC, a n d lo a d
t h e a p p r o p r i a t e b lo c k i f i t d o e s .
A f t e r i t h a s b e e n p a s s e d a num ber
o f K S 's , f g o a l w i l l t e r m i n a t e . I f i t h a s b e e n a b l e t o d i s c h a r g e i t s
d u t i e s , i t w i l l c h a n g e t h e SBC t o " g tm e r " ; i f i t h a s n 't , i t w i l l n o t
to u c h t h e SBC, b u t w i l l in f o r m m r i n t t h a t i t h a s f n i l e d .
M r in t w i l l ,
in tu rn , f a i l .
•
I f t h e SBC i s now " g tm e r " , g tm e r w i l l b e lo a d e d an d w i l l r u n e a c h tim e
a KS w i t h t h i s SBC field is loaded. When it i n
its
d u t i e s , g tm s r w i l l s e t t h e SBC t o
h a lt.
tu rn has
d is c h a rg e d
" f i n i s h " , a n d t h e s y s te m w i l l
*'fJ
T h u s , e a c h tim e a KS i s r e t r i e v e d w i t h i t s SBC f i e l d e q u a l t o t h e c u r r e n t
SBC, t h e a p p r o p r i a t e b l o c k w i l l b e lo a d e d , w i l l o p e r a t e o n t h e KS, and
w i l l t h e n t e r m i n a t e , c h a n g in g t h e c u r r e n t SBC i f
KS i s th e n r e t r i e v e d ..
THE DESIGN OF THE PROTOTYPE SYSTEM.
a p p ro p ria te .
T he n e x t
I t can be se en t h a t t h i s s tr a te g y
i s e q u i v a l e n t Vo u s i n g a num ber o f i n ­
d e p e n d e n t k n o w le d g e b a s e d s y s te m s w h ic h
in te r c o m m u n ic a te .
It
is ,
how­
e v e r , s lo w e r a s , i f we u s e t h e l a t t e r s t r a t e g y b e c a u s e we r e q u i r e c e r t a i n
o p e ra tio n s to
b e p e r f o r m e d b e f o r e o t h e r s , we may
p i p e l i n i n g t o i n c r e a s e t h e s y s te m ’ s s p e e d .
c a tio n
is s u e s
a re
s im p lif ie d ^
as
a ll
im p le m e n t
a
fo rm
o'
W ith t h i s s t r a t e g y , com m uni­
s y s te m
d a ta
b l o c k s , a n d a r e t h e n a v a i l a b l e t o a n y lo a d e d b l o c k .
a re h e ld
i n common
N o te t h a t t h e lo a d e r
a p p r o a c h i s n o t i d e a l l y s i m p l e ; t h e o r i g i n a l i n t e n t i o n w as t o
c o n s tru c t
a s i n g l e p r o g ra m , w i t h b lo c k s im p le m e n te d a s p r o c e d u r e s , a n d t h e p r e s e n t
T h is w as n o t p o s s i b l e , a s i t
le d t o code
t h a t e x c e e d e d t h e a v a i l a b l e c o d a s p a c e i n t h e c o m p u te r u s e d .
l o a d e r a *.>zm of case b l o c k .
T he d u t i e s
o f t h e b l o c k s w i l l now b e d e s c r i b e d .
T he f i r s t b l o c k in v o k e d b y t h e m a in s y s te m i s t h e i n i t s b l o c k , w h ic h e x ­
is ts
e n tire ly
a s a r e s u l t o f th e
im p l e m e n ta tio n
of
th e
s y s te m .
T he
m in ic o m p u te r o n w h ic h t h e s y s te m w as im p le m e n te d h a s a n a d d r e s s r a n g e o f
o n ly 32K w o r d s , w h ic h m eans t h a t a n y p ro g ra m t h a t c o m p ile s down t o g r e a t e r
t h a n 3 2 k w o rd s
( p ro g ra m a n d d a t a ) c a n n o t b e r u n .
T h e s y s te m w as im p le ­
m e n te d u s i n g a c t i v e d a t a s t r u c t u r e s , w h e re e a c h d a t a s t r u c t u r e i s im p le ­
m e n te d a s a p r o c e d u r e w i t h o p e r a t o r s
d e fin e d
as
p a ra m e te rs;
th e
d a ta
s t r u c t u r e s r e q u i r e i n i t i a l i s a t i o n , w h ic h w as im p le m e n te d a s an. o p e r a t o r ,
d e fin e d o n th e s t r u c t u r e .
T o w ard s t h e e n d of t h e p r o t o t y p i n g p h a s e , t h e
c o d e b e g a n t o c o m p ile down t o m o re t h a n 3 2 k , v i t h t h e r e s u l t t h a t t h e c o d e
had t o
b e s e g m e n te d , s o t h a t s e p a r a t e s e g a a n s s o f c o d e w e re lo a d e d i n t o
th e co d e S pace b y m r in t, and r u n .
I t w as f e l t t h a t i t w o u ld b e s im p le r
t o c o l l e c t a l l t h e d a t a - s t r u c t u r e p r o c e d u r e s i n t o a se g m e n t w h ose s i n g l e
d u ty w as t o
in itia lis e
th em ( t h e d a t a th e m s e lv e s b e in g k e p t i n common)
t h a n t o ad d c o d e t h a t w o u ld i n i t i a l i s e t h e s t r u c t u r e s t o t h e m a in p ro g ra m .
THE DESIGN OP THE PROTOTYPE SYSTEM.
T h is se g m e n t i s i n i t s , w h ic h i s u s e d o n ly t o i n i t i a l i s e th o s e s y s te m d a ta
s tr u c tu re s th a t re q u ir e i t .
T h e s e c o n d b l o c k t h a t a r i s e s f r o a t h e im p l e m e n ta tio n o f t h e s y s te m i s t h e
s t a r t b lo c k -, w h ic h w o u ld i n
KAB, a l b e i t a m in o r o n e .
f a c t be a p a r t o f th e g o a l s tr u c t u r e s e tu p
A t t h i s s t a g e , t h e s t a r t b lo c k i s u s e d o n l y t o
r e c o v e r t h e m a in s y s te m g o a l ; i t se em e d p o s s i b l e t h a t i t w o u ld b e r e q u i r e d
fo r
o th e r
fu n c tio n s ,
a n d w as t h u s im p le m e n te d a s a c o m p le te k n o w le d g e
b a sed b lo c k , a s th e i n f r a s t r u c t u r e e x is te d .
T h u s t h e o n ly tw o r u l e s
(in
t h i s c a s e th e y a r e m o re l i k e r u l e s o r la n g u a g e s t a t e m e n t s th a n K S 's ) t h a t
e x is t i n s t a r t a re :
•
A lw ays r e c o v e r t h e m a in s y s te m g o a l , and
•
a lw a y s c h a n g e SBC t o ( g o a l .
I t w i l l b e se en t h a t , in t h i s c a s e , th e o rd e r o f t h e i r
e x e c u tio n i s
im ­
p o r ta n t.
We h a v e d i s c u s s e d m r i n t a b o v e : t h e d u t i e s o f t h i s b lo c k a r e s im p le , b e in g
a n o th e r
b lo c k
th a t
has
a ris e n
fro m t h e im p l e m e n ta tio n o f t h e s y s te m .
M r in t i s r e q u i r e d t o r e t r i e v e fro m d i s k t h e r e p r e s e n t a t i o n o f t h e p i c t u r e ,
a n d t o s e t u p t h e s y s te m d a t a s t r u c t u r e s t h a t c o n t a i n t h i s r e p r e s e n t a t i o n .
It
th e n
p ro ce ss
lo a d s i n i t s
and,
when
o f l o a d i n g K S 's .
l o a d s t h e b lo c k i n d i c a t e d b y * ' .■
M r in t s t e p s th r o u g h th e
th a t
b lo c k
has
.te ra itiite d ,
b e g in s
th e
I f a n SBC f i e l d m a tc h e s t h e c u r r e n t SBC, i t
.
SBC h a s c h a n g e d t o " f i n i s h
•
• - r e n t SBC a n d w a i t s f o r i t t o c o m p le t e ,
if, o n e , an d c o n tin u e s t o d o s o u n t i l t h e
f
h i s s t a g e , i t p r o c e e d s t o s t e p th r o u g h
THE DESIGN OF THE PROTOTYPE SYSTEM.
th e K S 's w i t h o u t l o a d in g an y b lo c k , b u t t h i s
is
as a
re s u lt
of
m essy
c o d in g - i t c o u ld a s e a s i l y t e r m i n a t e .
We come now t o t h e i n t e r e s t i n g b l o c k s ; we w i l l d i s c u s s f g o a l f i r s t .
T he
d u ty o f t h i s b l o c k i s t o f a c t o r t h e m a in s y s te m g o a l i n t o p r i m i t i v e s ', an d
t o k e e p a r e c o r d o f t h e s t r u c t u r e o f t h e m ain s y s te m g o a l i n te rm s o f t h e s e
p r i m i t i v e s , u s i n g th e te c h n i q u e s we s h a l l d e s c r i b e b e lo w .
and s t r u c t u r a l d e s c r i p t i o n s
e r e n c e b y g tm e r .
The p r im it iv e
a r e k e p t i n s y s te m d a t a s t r u c t u r e s f o r r e f ­
T h e p r o c e s s b y w h ic h t h i s
b lo c k i s
in v o k e d m akes i t s
e x e c u t i o n p a r t i c u l a r l y s lo w , a s i t i s u n a b le t o s e a r c h t h e k n o w le d g e b a s e
a c t i v e l y , b u t m u st a c c e p t o r r e j e c t K S 's a s th e y a r e p a s s e d t o i t .
i s u n f o r t u n a t e , b u t t h e se g m e n t w as s im p ly e a s i e r t o
T h is
im p lem e n t i n t h i s
fa s h io n .
T he s e c o n d m a jo r b l o c k ,
g tm e r ,
is
re q u ire d
to
u se
s c r i p t i o n s p r o v id e d b y f g o a l an d t o m erg e r e g i o n s t o
th e
p rim itiv e
de­
fo rm t h e s e p r i m i ­
t i v e s , b y g e n e r a t i n g a n d t e s t i n g m e rg e s ( h e n c e i t s n a m e ) .
T h is b l o c k w i l l
th e n g e n e r a te g ro u p s o f r e g io n s a s c a n d id a te s f o r s t r u c t u r e , and w i l l t e s t
th em a g a i n s t t h e s t r u c t u r a l d e s c r i p t i o n s s t o r e d b y f g o a l .
te s t
has
been
u sed
e v a lu a te .s tru c tu re
is
h tr
i n te n d e d t o u s e a fo rm o f t h a t t e c h n i q u e ,
w i t h g r e a t e r s o p h i s t i c a t i o n t h a n t h i s b lo c k .
n e ig h b o u r
code
in fr a s tr u c tu re .
G e n e r a te arid
a s a t e c h n i q u e , a s t h e p a t h p l a n / a c t i o n p a th
m e rg in g
in f r a s tr u c tu re ,
_
as
a lb e it
T he b lo c k contains all t h e
w e ll
as
th e
KS e v a l u a t i o n
-V
By now t h e r e a d e r s h o u ld u n d e r s t a n d t h e h i g h - l e v e l d e s ig n o f t h e p r o t o ­
t y p e , a n d we s h a l l b e g in t o d i s c u s s th 6 m ore s o l i d d e s i g n i s s u e s .
THE DESIGN OF THE PROTOTYPE SYSTEM.
4.7
THE DESIGN OF THE DATA STRU CTURES.
In
a s y s te m o f t h i s
p lo y e d .
k in d ,
it
is
im p o r t a n t t h a t d a t a o r e c o r r e c t l y em­
T he s y s te m r e q u i r e s tw o m a jo r c a t e g o r i e s o f d a t a :
a r e em p lo y ed w i t h i n
a b l o c k w h ile i t
t h e n e x t tim e t h e b lo c k i s
in v o k e d .
S im p le d a t a
i s ru n n in g , and w i l l n o t be needed
S y stem d a t a
a r e r e q u ir e d e i t h e r by
m ore t h a n o n e b l o c k , o r b y o n e b l o c k o n m ore t h a n o n e o c c a s i o n .
We w i l l
d e s c r i b e o n ly t h e l a t t e r d a t a s t r u c t u r e s : t h e f o rm e r a r e d e s c r i b e d w h e re
n e c e s s a r y w h e re t h e b l o c k s a r e d e s c r i b e d /
We r e q u i r e t h e f o l l o w i n g s y s te m d a t a s t r u c t u r e s :
•
a s t r u c t u r e t o s t o r e a ctiuir _
. d e s c r ip tio n o f th e s tr u c t u r e o f
t h e m a in s y s te m g o a l ,
•
/
/
•
a
structure
into which s u b - g o a l s t h a t h a v e b e en f o u n d may b e
d e m o te d , a n d w h e re a r e c o r d o f s e t s
o f r e g i o n s t h a t m a tc h th em
may b e k e p t , a n d f i n a l l y ,
s t r u c t u r e s f o r t h e f g o a t b lo c k t o u s e a s a s c r a t c h p a d w h ile e x ­
p a n d in g a g o a l .
We u s e t h e te r m " s y s te m s u b - i '- a l" t o
re fe r to
a name ( o f e i t h e r a com­
p o s i t e o r a p r i m i t i v e ) t h a t is? n ;> * st o f t h e s t r u c t u r e o f t h e m ain s y s te m
g o a l , a n d h a s n o t b e e n e x p a n tir ^ .
firs t,
We w i l l d i s c u s s t h e
la tte r
s tru c tu re s
a s th e y a r e im p o r t a n t t o t h e d i s c u s s i o n o f t h e o t h e r s .
THE DESIGN OF THij PROTOTYPE SYSTEM.
48
T h e r e a r e tw o s t r u c t u r e s
u se d by f g o e l a s a s c r a t c h p a d : th e f i r s t ,
s y s te m c o m p o s ite l i s t ,
u se d t o k e e p t r a c k o f th e .
is
th e
sy s te m
s u b - g o a l s ,and
t h e s e c o n d , t h e s y s te m p r i m i t i v e 1 ' s t ,
i s u s e d t o k e e p t r a c k o f t h e s y s te m
p rim itiv e s .
lis t
T he
s y s te m
fo rw a rd - lin k e d l i s t ;
c o m p o s ite
move a n e le m e n t fro m a l i s t i n
in s p e c tio n
of
has
been
im p le m e n te d
as
a
t h i s s t r u c t u r e h a s b e e n u s e d a s i t i s s im p le t o r e ­
f la g s ,,
e way t h a t d o e s n o t
o rre s h u f f lin g
of
re c o rd s,
im p ed e
is
n e c e s s a r y .)
s e a r c h i n g . (No
T he
s t r u c t u r e c o n t a i n s t h e n am es o f a l l u n e x p a n d e d n am es i n t h e f a c t o r i n g o f
t h e m a in sy s te m g o a l .
T h u s , i t c o n t a i n s a l i s t o f t h e o b j e c t s w h o se d e ­
f i n i t i o n s t h e s y s te m s t i l l r e q u i r e s , an d i f a name d o e s n o t a p p e a r i n t h i s
lis t,
a KS d e s c r i b i n g a n o b j e c t o f t h a t name i s n o t o f i n t e r e s t t o f g o a l .
T he s y s te m p r i m i t i v e l i s t c o n t a i n s t h e d e s c r i p t i o n o f a l l p r i m i t i v e s w h ic h
h a v e b e e n e n c o u n te r e d i n f a c t o r i n g t h e m a in s y s te m g o a l , a n d w h o se n am es
w ere p a r t o f t h e s y s te m c o m p o s ite l i s t w hen th e y w e re e n c o u n te r e d l i s t i s n o t i n f a c t a lin k e d
lis t,
as i t
is
n o t se arch e d in t h i s
T h is
fo rm ,
b u t i s a s im p le ran d o m a c c e s s s t r u c t u r e .
T he s t r u c t u r e o f t h e m a in s y s te m g o a l i s s t o r e d i n t h e s t r u c t u r a l l i s t .
I t i s im p le m e n te d a s a f o r w a r d li n k e d l i s t ,
th r e e s t a c k s , a nd a lin k e d
f o r t h e s y s te m p r i m i t i v e
lis t.
T h e e le m e n ts t a k e t h e fo rm d e s c r i b e d b e ­
c a u s e o f t h e s t r u c t u r e o f t h e K S 's , w h o se
e v a lu a tio n i s
e a c h e le m e n t b e in g a s e t o f
l i s t b e in g c h o s e n t o r t h e same r e a s o n s g iv e n
s i m p l i f i e d by th e u s e
s y n ta x
i s such
o f s ta c k s .T h u s,
t h a t h a s listen r e c o r d e d m u st b e e v a l u a t e d f o r a
set
th a t
th e ir
when a s t r u c t u r e
o f r e g io n s ,
th e a p ­
p r o p r i a t e r e c o r d fro m t h e l i s t i s lo a d e d i n t o t h e e v a l u a t i n g p r o c e d u r e 's
s t a c k s , and t h e KS may b e e v a l u a t e d .
The n e e d f o r a s t r u c t u r e
i n t o w h ic h s u b - g o a l s t h a t h a v e b e e n f o u n d may
b e d e m o te d , a n d w h e re a r e c o r d o f s e t s o f r e g i o n s t h a t m atch th em 'm ay b e
THE DESIGN OF THE PROTOTYPE SYSTEM.
49
k e p t w i l l becom e a p p a r e n t when t h e a l g o r i t h m s b y w h ic h t h e b lo c k s o p e r a t e
a r e d e s c r i b e d ; we w i l l c a l l
it
th e s t r u c t u r a l p r im it iv e
lis t.
It
is
a
s im p le a r r a y o f r e c o r d s , w i t h a . " n e x t r e c o r d " p o i n t e r b e in g m a in ta in e d ,
as re c o rd s a re n o t d e le te d ,
becom e a p p a r e n t .
an d g a rb a g e c o l l e c t i o n o v e r h e a d s
E ac h r e c o r d c o n t a i n s a f i e l d
w ill
not
i n w h ic h t h e nam e o f th e
new p r i m i t i v e ( t h a t i s , e i t h e r a t r u e p r i m i t i v e , o r a c o m p o s ite t o w h ic h
a m a tc h h a s b e e r o b t a i n e d )
is
s to r e d ,
an d a f i e l d
i n w h ic h t h e n u m b ers
o f t h e r e g i o n s w h ic h m a tc h t h i s new p r i m i t i v e ' s d e s c r i p t i o n , a r e s t o r e d .
T he s y s te m d a t a s t r u c t u r e s a r e i m p o r ta n t t o t h e f u n c t i o n i n g o f t h e b lo c k
a lg o rith m s .
We h a v e d e s c r i b e d t h e s e s t r u c t u r e s ,
a lg o r i t h m s when we h a v e d i s c u s s e d a n o t h e r s u b j e c t
and w i l l d e s c r ib e th e
i m p o r ta n t
to
u n d e r­
s t a n d i n g t h e a l g o r i t h m s , t h e s y n t a x o f t h e K S 's .
THE SYNTAX AND STRUCTURE OF THE KNOWLEDGE S b U R C E S .
I t i s t r i t e t h a t a r e s t r i c t i v e a n d c a r e f u l l y - c h o s e n s y n ta x w i l l i n g e n e r a l
le a d to
a la n g u a g e w h ic h i s s im p le t o im p le m e n t.
T h e k n o w le d g e s o u r c e s
c a n b e t h o u g h t o f a s f o rm in g a s im p le la n g u a g e , b y w h ic h t h e
a p p e a ra n c e
o f o b j e c t s i n p i c t u r e s may b e d e s c r i b e d , s o we h a v e u s e d t h e a b o v e p r i n ­
c i p l e i n t h e i r d e s i g n t o s i m p l i f y t h e u s e o f t h e K S 's .
T h e i r s y n ta x h a s
b e e n d e s ig n e d w i t h s i m p l i c i t y o f im p l e m e n ta tio n b e in g t h e p r im e c o n s i d ­
e ra tio n .
A KS c o n t a i n s
f ie ld ,
an d
fo u r f ie ld s :
an
a c tio n
a n u m b er f i e l d ,
fie ld ,
in
THE DESIGN OF THE PROTOTYPE SYSTEM.
an
SBC f i e l d ,
th a t * o rd er.
We
a
have
c o n d itio n
used
th e
" c o n d i t i o n - a c t i o n " te r m in o lo g y h e r e f o r s i m p l i c i t y , an d a r e a w a re o f i t s
i n c o n s i s t e n c y w i t h t h e d i s l i k e o f t h e r u l e s te r m in o lo g y e x p r e s s e d a b o v e .
T he u s e o f t h e S»C f i e l d
h a s b e en e x p l a i n e d a b o v e : t h e n u m b er f i e l d
is
u s e d t o make t h e s y s te m e r r o r m e ssa g e s m o re m e a n in g f u l, s o t h a t t h e u s e r
knows i n w h ic h KS t h e s y s te m h a s f a i l e d .
T h e u s e s o f t h e c o n d i t i o n and
a c t i o n f i e l d s w i l l b e e x p l a i n e d b e lo w .
T h e co n d itio n f ie ld .
T he c o n d i t i o n f i e l d
c o n s is ts o f a s e t o f c o n d itio n s
lin k e d
by
lo g ic a l
o p e r a t o r s , f o rm in g a l o g i c a l e x p r e s s i o n , w h ic h im p l i e s t h e a c t i o n f i e l d .
'■Sti g e n e r a l ,
if
b lo c k t h e r u l e
th e
c o n d itio n
f ie ld
i s t r u e , t h e n , d e p e n d in g o n i n w h ic h
i s d e f in e d ,, e i t h e r t h e o b j e c t nam ed i n t h e a c t i o n
f ie ld
h a s b e e n f o u n d , o r t h e a c t i o n nam ed i n t h e a c t i o n f i e l d s h o u ld b e ta k e n .
T he common e x p e r t s y s te m p r a c t i c e
( F o r e x a m p le , i n t h e MYCIN system d e ­
s c r i b e d b y B u c h an a n a n d S h o r t l i f f e ' ( 1 9 8 4 ) i s t o p r o v i d e o n l y t h e l o g i c a l
o p e r a t o r AND i n t h e KS la n g u a g e .
T h u s , i f we w is h t o e x p r e s s t h a t i t w i l l
r a i n to m o rro w i f i t i s r a i n i n g t o d a y , o r i f i t i s o v e r c a s t t o d a y an d t h e
b a r o m e te r i s
f a llin g
(1 d o n o t v o u c h f o r t h i s K S ), we w o u ld b e o b lig e d
t o u s e tw o r u l e s , a n d s a y ;
•
•
IF i t
i s r a i n i n g t o d a y , THEN i t w i l l r a i n to m o rro w .
IF i t
i s o v e r c a s t t o d a y , AND t h e b a r o m e te r i s d r o p p in g , THEN i t
w i l l ' r a i n to m o rro w .
THE DESIGN OF THE PROTOTYPE SYSTEM.
The OR o p e r a t o r i s th e n a c h ie v e d b e c a u s e e i t h e r o n e , o r b o t h , o f t h e K S 's
may f i r e .
We h a v e n o t u s e d t h i s t e c h n i q u e b e c a u s e o f t h e o p a c i t y i t le n d s
t h e r u l e b a s e ( I t i s n o t e a s y t o s e e a t a. g la n c e j u s t w h at se g m e n t o f t h e
ru le base r e la te s
t o w h at c o n c lu s i o n .) , and b e c a u se p r o v is io n o f a f u l l
s e t o f o p e r a t o r s i s n o t s im p le i n t h e r u l e s t r u c t u r e u s e d .
We h a v e p r o v i d e d a s l o g i c a l o p e r a t o r s AND, OR, a n d NOT.
e n title d
to
o n l y tw o o p e r a n d s ,
AND a n d OR a r e
a n d NOT may h a v e o n ly o n e .
n a y b e e x p r e s s i o n s , o r s im p le c o n d i t i o n s .
T he o p e ra n d s
The o p e r a to r s a r e e x p re s s e d
i n p o s t - f i x n o t a t i o n , b e c a u s e i t i s e x tr e m e ly s im p le t o c o n s t r u c t a n i n ­
t e r p r e t e r f o r t h i s n o t a t i o n , o r i e n t e d a s i t i s to w a rd a s t a c k s t r u c t u r e .
N e s t i n g i s a llo w e d t o a d e p th o f 2 0 , t h e l i m i t b e i n g im p o sed f o r p r a c t i c a l
r e a s o n s o n ly .
We h a v e m e n tio n e d a b o v e t h e i m p o r t a n t s i m p l i f i c a t i o n t h a t h a s b e e n u s e d
in
th e c o n s tr u c tio n o f th e c o n d itio n
l i s t s y n ta x :
We d o n o t
a llo w
a
condition to be d e f i n e d b y t h e c o n d i t i o n l i s t o f o n e KS ( i . e . b y a p p e a r in g
in
its
if
t h e b lo c k t h a t u s e s t h e K S 's a t t e m p t s t o
a c tio n
fie ld )
and th e n t o
b e u s e d b y a n o th e r KS a s a c o n d i t i o n ,
e v a l u a t e " them
r a t h e r t h a n s t o r i n g th e m f o r s t r u c t u r a l i n f o r m a t i o n .
to
io m e d if - .'_ y
T h is i s e q u i v a l e n t
f o r b i d d i n g t h e u s e o f K S 's w h ic h r e q u i r e b a c k w a rd c h a i n i n g f o r t h v i r
e v a l u a t i o n , i f t h e b lo c k - i n w h ic h t h e y a p p e a r e v a l u a t e s K S 's im m e d ia te ly ,
and i s
a s i m p l i f i c a t i o n b e c a u s e w hen we e v a l u a t e K S 's , we m u st
v a ria b le s f o r c o n d itio n e v a lu a tio n .
k n o w le d g e
base,
d e fin e
I f we th e n w is h t o c h a i n t h r o u g h t h e
we m u st c a r r y t h e s e v a r i a b l e s .
We w i l l t h u s i n c u r a
c o n s id e r a b le o v e rh e a d t o no good p u r p o s e , a s , i f
a c o n d i t i o n may b e d e ­
f i n e d i n t h e f a s h i o n we a r e f o r b i d d i n g , e i t h e r i t m u st b e c a p a b le O f b e in g
re so lv e d ’in to
an e x p re s s io n ,
o r,
if
it
can n o t be re so lv e d ,
i t i s m ean­
i n g l e s s any h o w .■ T h u s we g a i n a t y p i n g o v e rh e a d i n e n t e r i n g K S 's b y lo s i n g
THE DESIGN OF THE PROTOTYPE SYSTEM.
a l a r g e c o d in g o v e r h e a d .
An im p l e m e n ta tio n i n
a la n g u a g e t h a t
a llo w s
r e c u r s io n w i l l n o t in c u r t h i s o v e rh e a d , as f a r as I can s e e .
The s i t u a t i o n
i s v e r y d i f f e r e n t when t h e b lo c k u s i n g t h e r u l e s u s e s them
to re c o rd s tr u c tu re :
in
th is
c a s e , t h e r u l e s a r e b e in g lo a d e d i n t o d a t a
s t r u c t u r e s f o r f u t u r e r e f e r e n c e , a n d when th e y a r e b e in g e v a l u a t e d , th e y
w i l l b e e v a l u a t e d i n a f o r w a r d - c h a i n i n g f a s h i o n , , t h e s y s te m h a v in g a l r e a d y
d e c id e d w h at c o n d i t i o n s a r e p r i m i t i v e , s o t h e p r o b le m o f r e c o r d i n g v a r i ­
a b le s i s n o t a s s e v e re .
A t t h e sam e ti m e , i n
th is
case i t
is
e s s e n tia l
t o a llo w b a c k w a rd r e f e r e n c e s , a n d t h e o v e rh e a d i n c u r r e d i s m ore a c c e p t a -
T h u s , f o r e x a m p le , t h e s y s te m o f K S 's show n b e lo w i s
i l l e g a l (I do n o t
v o u c h f o r t h e s e K S 's e i t h e r . ) :
•
I F i t i s o v e r c a s t AND t h e g a r d e n e r 's j o i n t s a r e aching,T H E N i t
i s l i k e l y to rain.
•
IF
it
is
lik e ly
to
r a i n AND t h e b a r o m e te r f a l l s , THEN i t
is
r a in in g .
■ B u t we may e a s i l y r e s o l v e t h e a b o v e s y s te m i n t o :
e
IF i t
i s o v e r c a s t , AND t h e i g a r d e n e r 's j o i n t s a r e a c h in g
AND t h e b a r o m e te r f a l l s , THEN i t i s r a i n i n g .
C o n d i tio n s may h a v e u p t o s e v e n o p e ra n d s ( a r e s u l t o f t h e im p l e m e n ta tio n ;
we h a v e n o t r e q u i r e d i n p r a c t i c e m ore th a n t h r e e ) , n o n e o f w h ic h may i t -
THE DESIGN OF THE PROTOTYPE SYSTEM.
. 53
s e l f be a c o n d itio n .
T h is r e s u l t s i n a s i m p l i f i c a t i o n
h a v e n o t im p lem e n te d v a r i a b l e - l e n g t h p a r a m e t e r l i s t s
a g a in f o r t h e s a k e o f s i m p l i c i t y .
of th t
code.
We
f o r th e c o n d itio n s ,
New c o n d i t i o n s a r e e a s i l y d e f i n e d .
T h e actio n fie ld .
T he a c t i o n f i e l d
ty p ic a lly
c o n ta in s
e i t h e r t h e name o f an a c t i o n ( w i th
p a r a m e t e r s ) , t h a t m u st b e p e r f o r m e d , o r t h e name o f a n o b j e c t t h a t
is
b e i n g d e s c r i b e d b y t h e K S, w i t h a p a r a m e t e r t o s a y w h e th e r t h e o b j e c t i s
c o m p o s ite o r p r i m i t i v e .
M u l t i p l e a c t i o n n am es a r e n o t a llo w e d ( f o r im ­
p le m e n t a t i o n s i m p l i c i t y ) , a n d m u l t i p l e o h '
* n am es w o u ld b e m e a n in g l e s s .
Exam ples o f ru le u s e .
To t h i s p o i n t i t h a s b e e n i m p l ie d t h a t t h e
to d e s c rib e o b je c ts
an d t o
same s t r u c t u r e h a s b e e n u s e d
im p ly a c t i o n s ; t h i s f a c u l t y i s a c h ie v e d b y a
s l i g h t l y d i f f e r e n t i n t e r p r e t a t i o n o f t h e s t a c k s g e n e r a te d b y t h e KS.
( n o tin g - th a t ttin d itio n and a c tio n f i e l d s a r e s e p a r a te d by a
T hus
and t h a t
a c o n d i t i o n p a r a m e t e r l i s t i s e n c lo s e d i n d iam o n d b r a c k e t s , t o s i m p l i f y
p a r s i n g t h e K S):
( ( ( ( (0N<LEG1, BODY> 0N<LEG2, B0DY>)A
ON<LEG3,BODY>)A 0N<LEG4,B0DY>)A
ON<BODY, IEAD>) A NEXT<HEAD,TRUNK>)A*ELEFNT<COMP>
THE DESIGN OF THE PROTOTYPE SYSTEM.
d e s c r i p t i o n o f an e l e p h a n t ;
s a r e s u l t o f th e
t h e name i s
s p e lle d
im p l e m e n ta tio n , nam es a r e
e le f n t
re q u ire d to
n s ix le tte r s .
T h e KS may b e d e s c r i b e d a s s a y i n g : a n o b j e c t , w i t h a s i n g l e b o d y s u p p o r te d
b y f o u r d i s t i n c t l e g s , 'a n d w i t h a h e a d o n t h a t b o d y , a n d a t r u n k a t t a c h e d
t o th e h e ad , i s an e le p h a n t.
p o s ite : th a t i s ,
th e p a r ts
I t s a y s a s w e l l t h a t a n e l e p h a n t i s a com­
i n te rm s o f w h ic h t h e e l e p h a n t i s d e s c r i b e d
m u st th e m s e lv e s b e f a c t o r e d i n t o p r i m i t i v e s , w h ic h a r e d e s c r i b e d i n te rm s
o f p ic tu re p ro p e rtie s .
N o tic e t h a t t h e n u m e r i c a l t a g s
appended t o
nam es o f t h e e l e p h a n t p a r t s d e s c r i b e t h e s e p a r t s a s b e in g d i s t i n c t .
e l e p h a n t m u st b y t h i s d e s c r i p t i o n h a v e f o u r l e g s .
b e f o r e t h e k n o w le d g e b a s e i s
th e
The
T he ta g s a re s tr ip p e d
search e d fo r th e p a r t 's d e s c r ip tio n , as i t
i s n o t a p p e a l i n g t o h a v e t o e n t e r K S 's d e s c r i b i n g I e g l , l e g 2 , l e g 3 , a n d s o
K S 's t h a t d e s c r i b e p r i m i t i v e s , d e s c r i b e th e m i n te r m s o f p i c t u r e
p ro p e rtie s .
le v e l
T h u s , a r o c k w o u ld b e d e s c r i b e d a s f o l l o w s :
((((SHAPE<DK> SIZE<DK>)A REFLEC<MEDIUM>)A TEXTUR<M£DIUM>)A
CONCAV<HIGH>)A*ROCK<PRIM>
T h is s a y s t h a t a r o c k h a s a n unknow n s h a p e a n d s i z e , a h i g h c o n c a v i t y (We
use
th is
te r m
to
mean
ti
rag g e d
o u tlin e .)
r e f l e c t a n c e , and t h a t i t i s a p r im it iv e .
lik e ly
tiv e ,
it
I t is
an d a medium t e x t u r e
and
a c c e p te d t h a t t h e r e a r e
t o b e p r o b le m s w i t h t h e in f o r m a t i o n g iv e n t o r e p r e s e n t t h e p r i m i ­
b u t a t th e p re se n t s ta te
o f im p l e m e n ta tio n i t
is
s u f f i c i e n t , and
i s o p e n t o e x p a n s io n .
THE DESIGN OF THE PROTOTYPE SYSTEM.
55
T he f o llo w in g KS e x p r e s s e s an a c t i o n :
(((((R0CK<R1,VERY> ROCK<CENTRE,Q+>)A (R0CK<R1,VERY> BACK<CENTRBvQ+>)A)0(.
R0CK<S1,VE8Y> ROGK<CENTRE,DK>)0 RNN1«<CENTRE,1>>A*WERGE<C£OTRE,R1>
n i l s r u l e s a y s t h a t i f t h e f i r s t r e g i o n i n a g iv e n r e g i o n ' s n e ig h b o u r c o d e
b e lo n g s s t r o n g l y t o t h e r o c k c l a s s , an d t h e g iv e n r e g i o n i t s e l f b e lo n g s
t o t h e r o c k c l a s s t o a d e g r e e o f q u i t e o r g r e a t e r , o r i f t h e g iv e n r e g i o n 's
f i r s t n e ig h b o u r b e lo n g s s t r o n g l y t o t h e r o c k c l a s s , an d t h e r e g i o n i t s e l f
b e lo n g s t o t h e b a c k g r o u n d c l a s s t o a d e g r e e o f q u i t e o r g r e a t e r , o r t h e
g iv e n r e g i o n ' s f i r s t n e ig h b o u r b e lo n g s s t r o n g l y t o t h e
ro c k c l a s s , and
t h e g iv e n r e g i o n b e lo n g s t o t h e r o c k c l a s s t o a d e g re e o f d o n 't know , an d
th e
g iv e n
r e g io n
has
o n ly
n e ig h b o u r s h o u ld b e m e rg e d .
one
n e ig h b o u r ,
th e n
th a t
re g io n
and
its
T h is i s t h e sam e a s s a y in g t h a t o n e may n o t
h a v e r o c k s w i t h h o l e s i n th e m .
An a l e r t r e a d e r w i l l h a v e n o t i c e d t h a t a s t h e n e ig h b o u r c o d e i s
it- i s
m e a n in g le s s t o
r e f e r to
th e
f i r s t r e g io n in i t .
c y c lic ,
W hat R1 i n f a c t
r e f e r s t o i s a l l p o s s i b l e f i r s t re g io n s in th e c o d e : t h u s , by r e f e r r i n g
t o R l a n d R3
re g io n .
we may r e f e r t o a n y tw o r e g i o n s i n t h e c o d e o f f s e t b y o n e
We w i l l d i s c u s s t h i s p o i n t i n g r e a t e r d e p th b e lo w .
I t i s o b v io u s t h a t t h e KS f o r m a t i s e x tr e m e ly o p a q u e ; t h e s y s te m h a s n o t
i n a n y way b e e n d e s ig n e d f o r a n u n s k i l l e d o p e r a t o r .
It
is
of
c o u rs e
p o s s i b l e t o a t t a c h a n a t u r a l la n g u a g e i n t e r f a c e t o t h e p r e s e n t s im p le r u l e
e d i t o r , b u t t h i s w as f e l t t o b e f a r b e y o n d t h e s c o p e o f t h i s p r o t o t y p e .
THE DESIGN OF THE PROTOTYPE SYSTEM.
We h a v e d i s c u s s e r
t h e s y s te m b lo c k d ia g :
an d t h e s y n t a x a n d s t r u c t u r e o f t h e K S 's
r l t h m s u s e d b y t h e b lo c k s .
THE ALGORITHMS USED W I T H I N T H E B L O C K S .
T he a l g o r i t h m s u s e d b y t h e b l o c k s f g o a l a n d g tm e r t o d i s c h a r g e t h e i r d u ­
tie s
w ill
n o t b e p re s e n te d
fo rm a lly , b u t w i l l be d e s c rib e d i n d e t a i l .
T h ey a r e e s s e n t i a l l y f a i r l y s i m p l e , b u t a f o r m a l p r e s e n t a t i o n w o u ld i n ­
v o l v e a m ass o f u n im p o r ta n t d e t a i l , a n d .w o u ld c lo u d t h e i s s u e .
d e s c r i b e how t h e s y s te m e v a l u a t e s a KS, how i t
g o a l and s t o r e s
its
We w i l l
f a c t o r s t h e m a in sy s te m
s t r u c t u r a l d e s c r i p t i o n , how i t
g e n e r a te s and t e s t s
m e rg e s , an d how i t f o r w a r d c h a i n s t o t h e m a in s y s te m g o a l .
T he l a s t a l ­
g o r ith m h a s n o t y e t b e e n im p le m e n te d , a n d a s s u c h i s n o t d e f i n i t e l y f r e e
o f p r o b le m s .
T h e b lo c k s b a s e d o n t h e o t h e r a lg o r i th m s w o rk , b u t t h i s d o e s
n o t g u a ra n te e th e c o r r e c tn e s s o f th e a lg o rith m s .
T h ese a lg o rith m s a r e n o t t r u l y f a i l - s a f e ,
im p le m e n te d
code
a re
tr a p p e d
and
e r r o r - h a n d l i n g f a c i l i t i e s w hen we
b u t e r r o r s e n c o u n te r e d b y t h e
r e p o rte d ;
d e s c rib e
th e
we
w ill
d e s c rib e
im p l e m e n ta tio n
of
th e
th e
p ro to ty p e .
THE DESIGN OF THE PROTOTYPE SYSTEM.
i s
,
■i ■
r.7
How th e p r o to ty p e e v a lu a te s a KS.
We h a v e d i s c u s s e d a b o v e t h e s y n ta x a n d s t r u c t u r e o f t h e K S 's an d h a v e s a i d
how t h e y h a v e b e e n o r d e a t e d t o s i m p l i f y e v a l u a t i o n o f t h e KS.
We w i l l
show h e r e j u s t how th e y s i m p l i f y t h i s p r o c e s s , an d how t h e s t a c k - o r i e n t e d
s t r u c t u r e o f t h e K S 's s i m p l i f i e s t h e i r im p l e m e n ta tio n .
T h e p r o c e s s o f KS e v a l u a t i o n b o i l s down t o f o u r s t e p s :
•
T h e K S 's m u st b e d e co m p o se d i n t o t h e i r s t a c k r e p r e s e n t a t i o n .
•
T h e c o n d i t i o n o p e r a n d s f o r e a c h c o n d i t i o n m u st b e f i x e d .
•
T h e c o n d i t i o n s th e m s e lv e s m u st b e e v a l u a t e d .
•
F in a lly , a l l
t h e v a l u e s o f t h e c o n d i t i o n s m u st b e co m b in ed t o fo rm
t h e v a l u e o f t h e KS.
D e c o m p o s in g K S ’s :
T he KS i s
T h i s p r o c e s s i s im p le m e n te d b y a s im p le s t a t e m a c h in e .
r e a d fro m b a c k t o
fro n t
( th is
le a d s t o t h e s t a c k s b e in g i n a
m o re c o n v e n ie n t o r d e r , a n d t h e n am es a r e e a s i l y r e v e r s e d )
e ra to r ( " (" ,
o p e ra to r s ta c k .
an d i f
an o p ­
"A11, " 0 " o r "N ") i s e n c o u n te r e d , i t i s p u sh e d i n t o t h e
T h e s t a r t o f a c o n d i t i o n f i d 'i d w i l l b e i n d i c a t e d by
(rem em b er we a r e r e a d i n g t h e KS b a c k w a r d s , an d n o t i c e now t h e r e a s o n f o r
u s i n g d iam o n d b r a c k e t s f o r d e l i m i t i n g c o n d i t i o n p a r a m e t e r l i s t s ) . ' A f t e r
th is
we
le tte rs
e x p ec t up t o
s ix
le tte rs ,
fo rm t h e name o f t h e
r e v e rs e d and pu sh ed i n t o
f o llo w e d b y a " ,
" or a
th e s e
l a s t o f t h e c o n d i t i o n o p e r a n d s , w h ic h
t h e c o n d i t i o n o p e ra n d s t a c k .
THE DESIGN OF THE PROTOTYPE SYSTEM.
is
I f th e d e lim ite r
38
e n c o u n tp
' i s a " , " th e n we c o n tin u e t o f i n d an d p u sh c o n d i t i o n o p e ra n d s
u n t i l we e r u v n ' . e r a
We d o n o t s t o r e
th e
d iam o n d b r a c k e t s o r th e
com m as, n o r d o we d e l i m i t t h e o p e ra n d s i n t h e s t a c k a s b e lo n g in g t o s e p ­
a r a te c o n d itio n s .
im p o s ed a t t h i s
N e t s t h a t t h e l i m i t o f s e v e n c o n d i t i o n o p e ra n d s i s n o t
p o in t.
A KS e x c e e d in g t h i s
lim it,
t a i n i n g m ore o r f e w e r o p e r a n d s th a n a r e d e f i n e d
o r a c o n d itio n con­
fo r i t ,
w ill cau se
an
e r r o r b a s e d o n t h e d i s a r r a y o f t h e c o n d i t i o n o p e ra n d s t a c k , a s t h e r e W ill
be
m o re , o r f e w e r o p e ra n d s th a n r e q u i r e d
e n fo r c e s ty p in g on i t s
in th e s ta c k .
o p e r a n d s , t h e e r r o r s h o u ld
be
As t h e s y s te m
re p o rte d
w hen
a
c o n d i t i o n r e c e i v e s a n o p e ra n d n o t i n te n d e d f o r i t .
When a "< " h a s b e e n e n c o u n te r e d , t h e f o llo w i n g c h a r a c t e r s , t o a maximum
o f s i x , fo rm t h e c o n d i t i o n n am e, w h ic h i s
t h e c o n d itio n s ta c k .
fie ld
h a s b e e n d e co m p o se d .
not "<" n o r
r e v e r s e d an d th e n p u s h e d o n to
T h e p r o c e s s c o n tin u e s u n t i l t h e e n t i r e KS c o n d i t i o n
We c a t c h u n m a tc h e d p a r e n t h e s e s ( " ( " o r
w h ic h a r e u s e d o n ly t o
d e lim it th e c o n d itio n p a ra m e te r
a r e a ) b y s e t t i n g a c o u n t e r t o z e r o , in c r e m e n tin g i t e a c h tim e e left p a ­
r e n th e s is is
e n c o u n te r e d ,
a n d d e c r e m e n tin g i t
f o r a r ig h t
p a re n th e s is .
I f w hen t h e c o n d i t i o n f i e l d h a s b e e n f u l l y d eco m p o se d t h e c o u n te r i s n o t
a g a i n z e r o , we h a v e a n e r r o r .
E v a lu a tin g t h e c o n d itio n s . :
The n e x t p r o c e s s i s
t h a t o f e v a lu a tin g a l l
t h e c o n d i t i o n s , w h ile r e t a i n i n g t h e i r o r d e r i n t h e s t a c k ,
We d o t h i s b y
p o p p in g e a c h c o n d i t i o n fro m t h e c o n d i t i o n s t a c k u n t i l t h e s t a c k r e t u r n s
a " fa il"
s t a t u s , in d ic a t in g t h a t i s
em pty ( t h e r e m u st b e a t l e a s t
one
c o n d i t i o n ) , e v a l u a t i n g e a c h c o n d i t i o n a s i t i s p o p p e d , and th e n e n q u e u e in g
th e v a lu e s so t h a t th e o r d e r i s r e ta in e d .
THE DESl'GN OF THE PROTOTYPE SYSTEM.
Wlion
a
c o n d itio n
is
popped
fro m t h e c o n d i t i o n
s ta c k ,
th e a p p ro p ria te
c o n d i t i o n e v a l u a t i o n p r o c e d u r e i s i n v o k e d , w h ic h th e n p o p s t h e num ber o f
c o n d itio n o p e ra n d s
s ta c k .
fix e d
f o r t h a t p r o c e d u r e fro m t h e c o n d i t i o n
o p e ra n d
T h e s e o p e r a n d s m u st t h e n b e f i x e d : t h e p r o c e s s i s d e s c r i b e d b e lo w .
O nce t h e c o n d i t i o n o p e r a n d s h a v e b e e n p o p p e d a n d f i x e d , t h e c o n d i t i o n m u st
b e e v a lu a te d .
T h is i s a p r o c e s s t h a t r a t h e r d e p e n d s o n how t h e c o n d i t i o n
i s d e f i n e d , b u t we w i l l re m a rk t h a t when a c o n d i t i o n f i x e s a n o p e ra n d an d
r e c e iv e s a v e c t o r ,
i t w i l l t h e n u s e a n y s c a l a r o p e ra n d a s a v e c t o r , b y
s im p ly f i l l i n g a v e c t o r o f t h e g iv e n l e n g t h w i t h t h e s c a l a r v a l u e ,
and
r e t u r n a v e c t o r r e s u l t (w h ic h w i l l c o n t a i n a s i t s n ' t h e le m e n t t h e r e s u l t
o f t h e c o n d itio n o p e r a tin g on t h e o p e ra n d s in th e n 't h p la c e o f th e op ­
e ra n d v e c t o r s ) .
T he r e s u l t r e t u r n e d b y a c o n d i t i o n b e i n t h e r a n g e 0
-
1 ( to ta lly fa ls e to to ta lly tr u e , re s p e c tiv e ly ) .
As t h e c o n d i t i o n s r e t u r n t h e i r
r e s u l t s , th e r e s u l t s a re enqueued, rea d y
f o r t h e KS evaluator, w h ic h u s e s only them a n d t h e c o n t e n t s o f t h e o p e r ­
a t o r s t a c k t o e v a l u a t e t h e KS.
F ixing c o n d itio n o p e r a n d s .:
v a lu e a n d a t y p e .
n o t m e r e ly o f
To f i x
a n o p e r a n d , we a s s o c i a t e w ith i t
a
As t r u e v a r i a b l e s a r e n o t u s e d h e r e , i t i s a q u e s t i o n ,
" in d in g t h e a p p r o p r i a t e p a r t o f a v a r i a b l e
sp a ce and r e ­
t r i e v i n g a v a l u e , b u t o f d e te r m in i n g t h e ty p e o f t h e o p e r a n d , a n d e i t h e r
a s s o c i a tin g w ith i t
th e
f i x e d v a lu e a s s o c i a t e d w i t h t h a t name
in
th a t
t y p e , o r r e t r i e v i n g t h e r e g i o n n u m b e r, s e t o f r e g i o n s o r num ber a s s o c i a t e d
w ith
it.
We h a v e f o u r t y p e s a t p r e s e n t :
n u m b e rs , a n d s y s te m p r i m i t i v e s .
THE' DESIGN OF THE PROTOTYPE SYSTEM.
q u a lifie rs ,
r e g i o n n u m b e rs ,
A q u a l i f i e r i s a member o f t h e s e t
QN- 1
, and i s
VERY, QUITE, DK, QUITEN, VERYN, Q+,
a s s o c i a te d w ith a c o n s ta n t v a lu e o r ra n g e o f v a lu e s , d e ­
p e n d in g on w h a t <ji
w ith in th e e o n d itio .i,
tie r i t i s .
f o r th e
T h is v a lu e i s t h e n com pared t o a v a lfie
c o n d itio n t o
r e tu r n
a one ( t o t a l l y
tru e )
o r a z e r o ( t o t a l l y f a l s a ) v a lu e .
A r e g i o n n u m b er i s d e f i n e d
w i t h w h ic h t h e ' s y s te m i s
e i t h e r a s CENTRE, w h ic h r e c o v e r s t h e r e g i o n
d e a l i n g a t t h e tim e i t
is
e n c o u n te r e d , o r
an
i n t e g e r w i t h a n "R " p r e c e d i n g t h e f i r s t d i g i t , w h ic h w i l l r e c o v e r e i t h e r
t h e e n t i r e c o d e , s t a r t i n g w i t h t h e r e g i o n o c c u p y in g t h e p l a c e i n t h e c o d e
referred
t o b y t h e n u m b e r, o r a f l a g
t h a t w i l l f o r c e th e c o n d itio n
to
e v a l u a t e a s z e r o ( t o t a l l y f a l s e ) , if t h e n u m b er i s g r e a t e r th a n t h e le n g t h
o f th e code.
As t h e c o d e s t a r t s w i t h
m e a n in g le s s t o
r e f e r to
an
a rb itra ry
n e ig h b o u r ,
it
is
t h e f i r s t r e g i o n i n t h e c o d e , b u t a s t h e s y s te m
d o e s n o t t h e r e a f t e r r o t a t e t h e c o d e , we c a n , u s i n g t h i s sc h em e , r e f e r f o r
e x am p le t o
a ll
p a irs
of
n e ig h b o u r s
se p a ra te d
n e ig h b o u r s , w h ic h we w o u ld e x p r e s s a s R1 a n d R 4.
by
at
le a s t
tw o
o th e r
T h is l e a d s t o t h e p o i n t
t h a t c o n d i t i o n s o f t e n r e t u r n v e c t o r , r a t h e r t h a n s c a l a r v a l u e s , a s th e y
QUITEN => q u i t e I
QN- => q u i t e n o t o r l e s s
THE DESIGN OF THE PROTOTYPE SYSTEM.
a rc e v a lu a te d
f o r a s e t o f r e g io n s
fro m t h e n e ig h b o u r c o d e .
At
th is
s t a g e , t h e v e c t o r s w i l l a lw a y s h a v e t h e same l e n g t h , a s we a r e d o n o t y e t
ad low m o re, th a n o n e r e g i t n ' s
n e ig h b o u r c o d e t o
be r e f e r re d to
i n a KS:
t h e scalars a r e th e n e x p a n d e d t o a v e c t o r w ith t h i s l e n g t h , a n d t h e w h o le
KS e v a l u a t e d f o r c o n d i t i o n s w h ic h a r e v e c t o r s .
T h e KS e v a l u a t o r
w ill
r e t u r n e i t h e r t h e v e c t o r v a l u e o f t h e KS, o r t h e maximum s c a l a r v a l u e ,
d e p e n d in g o n how i t i s
in v o k e d .
T h e n u m b er ty p e c o n s i s t s o f i n t e g e r s w i t h o u t p r e f i x e s , w h ic h a r e e v a l u a t e d
a s t h e i n t e g e r g i v e n ; t h e s e o p e ra n d s a llo w t h e i n s p e c t i o n o f , f o r e x a m p le ,
t h e n u m b er o f n e ig h b o u r s a g iv e n r e g i o n h a s .
I t m u st b e s e e n t h a t a s y s te m p r i m i t i v e i s i n n o w ay t h e sam e a s a n a b ­
s o lu te p r im it iv e
( f o r w h ic h we h a v e u s e d t h e w o rd p r i m i t i v e ) , a s a n a b ­
s o l u t e p r i m i t i v e i s d e f i n e d b y t h e r u l e b a s e , a n d a s y s te m p r i m i t i v e may
b e e i t h e r a n a b s o l u t e p r i m i t i v e , o r a c o m p o s ite f o r w h ic h a m a tc h h a s b e e n
fo u n d .
A s y s te m p r i m i t i v e o p e ra n d r e f e r s t o t h e s e t o f m a tc h e s t h a t h a v e
b e e n f o u n d f o r t h e s y s te m p r i m i t i v e nam ed a s t h e o p e r a n d .
a lg o r i t h m f o r c h a i n i n g f o r w a r d t o
As t h e p r o p o s e d
a c h ie v e t h e m a in s y s te m g o a l h a s n o t
b e e n .im p le m e n te d , an d h e n c e n o c o n d i t i o n r e q u i r e s sy s te m p r i m i t i v e o p e r ­
a n d s, th e d e t a i l s o f t h e i r u se have n o t been f u l l y te s te d .
T he i n t e n t i o n
i s , h o w e v e r, a s f o l l o w s :
S y stem p r i m i t i v e
o p e ra n d s w i l l r e f e r t o s e t s o f r e g i o n s , a n d t h e c o n d i­
t i o n s u s i n g t h e s e o p e ra n d s w i l l h a v e t o b e h a n d le d a s s u c h .
c o n d itio n w ill hove to
th e o p e ra n d s s p e c if ie d .
to t y p e
s im p ly
by
b e e v a lu a te d
fo r a l l a v a ila b le
T hus
th e
a lte r n a tiv e s
fo r
I t w as i n te n d e d t o im p le m e n t t h i s
s e ttin g
f b r th e p ro ­
u p a s t r u c t u r e w h ic h w o u ld h o ld t h e c u r r e n t
THE DESIGN OF THE PROTOTYPE SYSTEM.
62
c a n d i d a t e s f o r a l l s y s te m p r i m i t i v e s r e f e r r e d t o w i t h i n a KS, l o a d i n g t h a t
KS fro m t h e s y s te m s t r u c t u r e l i s t , a n d e v a l u a t i n g t h e KS.
T h is p r o c e s s
w o u ld b e p e r f o r m e d f o r e a c h l e g a l s e t o f c u r r e n t c a n d td s . t e s .
E v a lu a tin g th e r u l e . :
The r u l e e v a lu a to r u s e s a te c h n iq u e t h a t i s v e s te d
i n o u r i n s i s t e n c e o n p o s t - f i x n o t a t i o n an d a d e f i n e d n u m b er o f o p e ra n d s
fo r each lo g ic a l o p e ra to r .
We w i l l d i s c u s s t h i s i n te rm s o f s c a l a r s o n l y ,
a lth o u g h t h e t e c h n i q u e e x te n d s v e r y s im p ly t o
f o rm in g
o p e ra tio n s
on
th e v e c to rs
r e t r i e v i n g th e m a s v e c t o r s .
v e c t o r s , m e r e ly
by
p e r­
e le m e n t- b y - e le m e n t , a n d s t o r i n g a n d
( E le m e n ts i n v e c t o r s w i l l n o t h a v e t o
be
s t o r e d i n d i f f e r e n t p a r t s o f t h e te m p o r a r y d a t a s t r u c t u r e u s e d : t h a t i s ,
s t o r a g e an d r e t r i e v a l n e e d n o t b e p e rf o rm e d e le r a e n t - b y - e l e m e n t . )
We w i l l
f i r s t d is c u s s th e o p e ra tio n s u se d .
When t h e t r u t h v a l u e i s d e f i n e d a s a member o f a r a n g e o f n u m b e rs , r a t h e r
th a n a s o n e o f tw o , we m u st d e f i n e f u n c t i o n s o n i t t o r e p r e s e n t t h e l o g ­
ic a l o p e ra to rs .
A s we h a v e a c c e p te d D u b o is a n d P r u d e 's
m u st a c c e p t t h e i r f u n c t i o n s .
s y s te m ( ) , we
T h u s , g iv e n tw o p r o p o s i t i o n s , T a n d R , we
THE DESIGN OF THE PROTOTYPE SYSTEM.
( ( T R) OR) TRUTHVALUE = MAX((T)TRUTHVALUE,(R)T«UTHVALOE)
T h e t r u t h v a l u e o f (T R)AND a s :
( ( T R)AND)TRUTHVALUE = MIN((T)TRUTHVALUE,(R)TRUTHVALUE)
T h e t r u t h v a l u e o f (T)NOT a s :
( (T)NOT)TRUTHVAIAJE = 1 - (T)TRUTHVALUE
And f o r a n y p r o p o s i t i o n S ,
0 <= (S)TRUTHVALUE <= 1
u s in g p o s t- f ix n o ta tio n f o r ty p e s e t tin g s im p lic ity .
(F o r f u r t h e r i n f o r ­
m a tio n , a n d a j u s t i f i c a t i o n o f t h e s e f u n c t i o n s , s e e t h e r e f e r e n c e . )
T he a l g o r i t h m u s e d t o
on
th e s ta c k o r d e r ,
e v a l u a t e t h e KS i s
t h e n q u i t e s im p le ,
and r e l i e s
a n d o n c a r e f u l u s e o f - a te m p o r a r y d a t d s t r u c t u r e .
T h e d a t a s t r u c t u r e i s a tw o - d im e n S io n a l a r r a y , w h ic h i s tw o e le m e n ts w id e ,
a n d w h o se d e p th d e f i n e s t h e l i m i t t o w h ic h u n m a tc h e d l e f t p a r e n t h e s e s may
b e e n c o u n te r e d i n t h e p r o c e s s o f e v a l u a t i o n .
(T h at i s ,
i f we g n c o u n 'jd r
20 l e f t p a r e n t h e s e s d u r i n g o u r e v a l u a t i o n b e fo re we s e e a r i g h t p a r e n ­
t h e s i s , t h e s t r u c t u r e m u st b e 20 d e e p ) .
THE DESIGN OF THE PROTOTYPE SYSTEM.
The e v a lu a tio n p ro c e e d s a s f o l -
•
T h e te m p o r a r y d a t a s t r u c t u r e ( c a l l e d te m p ) i s i n i t i a l i s e d t o a l l
dummy v a l u e s , a n d t h e b r a c k e t c o u n t e r i s s e t t o z e r o .
T he f i r s t
o c c u p a n t o f t h e o p e r a t o r s ta c k " " is now p o p p e d .
•
I f t h e o p e r a t o r s t a c k i s e m p ty , t h e r e c a n b e o n ly o n e c o n d i t i o n ,
and th e v a lu e o f th e c o n d itio n a t th e h e ad o f th e queue i s th e
v a lu e o f th e r u l e .
•
I f a " ( " i s p o p p e d , t h e b r a c k e t c o u n t e r in c r e m e n ts , a n d t h e n e x t
o p e ra to r i s popped.
•
I f f o llo w in g a
a " ) " i s p o p p e d , a n d t h e n an o p e r a t o r to k e n
( e i t h e r " A " ," 0 " o r " N " ) , we know t h a t we may d e q u e u e e i t h e r o n e
( f o r "N ") o r tw o ( f o r "A" o r ”0 " ) v a l u e s , an d co m b in e th em a c ­
c o r d in g t o
th e o p e ra to r popped.
T he r e s u l t a n t v a l u e
is
th e n
s to r e d .
•
To s t o r e a v a l u e , we i n s p e c t t h e f i r s t l o c a t i o n i n te m p a t t h e
c u r r e n t b r a c k e t d e p th ;
in s p e c t th e seco n d .
i f th is
c o n t a i n s a non-dummy v a l u e ,
we
I f n e i t h e r c o n t a i n s a dummy v a l u e , s o m e th in g
h a s g o n e w ro n g , a n d t h e te c h n i q u e f a i l s .
I f th e f i r s t c o n ta in s
a dummy, we s t o r e o u r v a l u e i n i t ; i f i t d o e s n ’t , a n d t h e se c o n d
d o e s , we s t o r e o u r v a lu e i n t h e s e c o n d .
•
I f i n s t e a d o f p o p p in g t h e se q u e r .. s
o n ly " ) "
" < o p e ra to r> " ,
" < o p e r a t o r > " , we p o p
im p l y in g t h a t a t l e a s t o n e o p e r a t o r h a s
b e e n e v a l u a t e d , we m u st i n s p e c t tem p f o r o p e r a n d s .
I f th e f i r s t '
l o c a t i o n a t t h e c u r r e n t b r a c k e t d e p th p l u s o n e , c o n t a i n s a dummy
. v a l u e , we f a i l .
I f t h i s lo c a tio n c o n ta in s a r e a l v a lu e , b u t th e
THE DESIGN OF THE PROTOTYPE SYSTEM.
s e c o n d l o c a t i o n a t t h i s d e p th c o n t a i n s s dummy v a l u e , « e u s e t b s
v a lu e i n t h e f i r s t l o c a t i o n an d a s i n g l e d e q u e u e d v a lu e
( i f we
n e ed a se co n d v a lu e ) a s th e o p e ra n d s, and e v a lu a te th e o p e ra to r
and
s to r e
t h e r e s u l t in t h e fashion d e s c r i b e d a b o v e .
second lo c a tio n
a ls o c o n ta in s
a r e a l v a lu e ,
I f th e
t h e tw o v a l u e s i n
tem p a r e u s e d a s o p e ra n d s f o r t h e o p e r a t o r ( i f t h e o p e r a t o r i s
a tw o - o p e r a n d o p e r a t o r : i f i t i s a o n e -o p e ra n d o p e r a t o r , we m u st
f a i l ) a nd th e r e s u l t s a re s to r e d a s d e s c rib e d .
*
We c o n t i n u e t h i s p r o c e s s u n t i l t h e o p e r a t o r s t a c k i s
u n til th e
a lg o rith m
in s u ffic ie n t
qu eued
fa ils
e m p ty , o r
a s a r e s u l t o f an a n o m alo u s tem p o r
o p e ra n d s,
c a u s e d b y KS s y n ta x e r r o r s .
b o th
of
w h ic h
c o n d itio n s
a re
We w i l l th e n f i n d t h e v a l u e o f t h e
KS i n t h e f i r s t l o c a t i o n i n tem p a t b r a c k e t l e v e l z e r o .
A lth o u g h t h e a l g o r i t h m s o u n d s c o m p le x , f i g u r e 8 w i l l i n d i c a t e how sim p le
it
r e a lly
is .
Its
s u s c e s s i s b a s e d on t h e s t r u c t u r e o f t h e K S 's , w h ic h
w as c a r e f u l l y d e s ig n e d f o r a n a lg o r i t h m o f t h i s t y p e .
As lo n g a s t h e KS
s y n ta x i s c o r r e c t , t h e a lg o r i t h m w o r k s .
We h a v e d e s c r i b e d i n
d e t a i l how t h e s y s te m e v a l u a t e s a K S; we w i l l now
d i s c u s s how t h e s y s te m f a c t o r s a g o a l .
THE DESIGN OF THE PROTOTYPE SYSTEM.
((Cl
c :1)/' (V i
We h a v e now s e e n ( ) A .
X h u s, we d e q u e u e tw o o p e r a n d s
a n d fo rm V I.V 2 , w h ic h g o e s i n t o th e f i r s t p l a c e i n
tem p a t l e v e l 3 .
FIGURE
To I l l u s t r a t e E v a l u a t i n g a KS,
i l l ) ' 6: ),u|. (
St.0)1 7 : pi,]) )
We h a v e s e e n ( )A , a n d a r e a s b e f o r e ^ e n t i t l e d
S tep 8 : _pop A
t o dequi
tw o o p e r a n d s , a n d fo rm V 3 .V 4 , w h ic h g o e s I n t o th e
s e c o n d l o c a t i o n a t l e v e l 3 i n te m p .
( t h e q u e u e i s e m p ty )
O p e ra to r s ta c k
S te p 9 : pop )
B r ■= 2
S te p 1 0 : pop O
He l a v e p o p p e d ) 0 , a n d m u s t I n s p e c t tem p a t a l e v e l
o n e d e e p e r th a n th e b r a c k e t d e p th t o f i n d o u t w h at
s h o u ld b e d o n e a b o u t o p e r a n d s ,
We f i n d b o t h l o c a t i o n s
o c c u p i e d , a n d h e n c e m u s t u s e t h e i r o c c u p a n t s . We
fo rm V I.V 2 + V 3 .V 4 , a n d s t o r e i t i n th e f i r s t l o c a t i o n
in
V l.V 2 -tV 3 .V 4
S te p 1 1
z pop )
tem p a t t h e b r a c k e t d e p th .
D
Br = 1
S te p 12 : p o p N
‘
A nd a g a i n , we m u st i n s p e c t tem p f o r o p e r a n d s .
T h ere i s
o n e , a n d we r e q u i r e o n ly o n e , s o we fo rm V 1.V 2+V 3.V 4, and
p la c e i t
i n t h e f i r s t l o c a t i o n a t th e b r a c k e t d e p th .
S te p 13 : T he o p e r a t o r s t a c k i s e m p ty , s o we i n s p e c t te m p , a .id f i n o u r r e s u l t
w h e re we e x p e c t e d i t .
FIGURE 8 :
B r i s a g a i n 1 , s o we may t e r m i n a t e " h a p p y " -
To I l l u s t r a t e E v a l u a t i n g a K S.
P a rt 2
How the prototype factors a goal.
T h e p r o t o t y p e u s e s some te c h n i q u e s p o p u la r i n t h e c u r r e n t A r t i f i c i a l I n ­
te llig e n c e lit e r a t u r e :
th is
b lo c k ,
f o r e x a m p le , u s e s a te c h n i q u e
known
a s b a c k w a rd c h a i n i n g , w h ic h r e a l l y d e s e r v e s d e s c r i p t i o n o n l y b e c a u s e t h e
la n g u a g e i n w h ic h t h e p r o t o t y p e w as im p le m e n te d d o e s n o t a llo w r e c u r s i o n ,
an d some i n g e n u i t y w as r e q u i r e d t o g e t t h e p r o t o t y p e t o
c h a i n b a c k w a rd .
We w i l l b r i e f l y d e s c r i b e t h e a l g o r i t h m t h a t t h e s y s te m u s e s .
When we r e q u i r e t h e s y s te m t o f a c t o r a g o a l , we a r e r e q u i r i n g i t t o r e c o r d
a ll
K S 's
n e ce ssary
to
d e s c rib e
th e
p r im it iv e s and c o n d itio n s on p r im it iv e s .
goal
c o m p le t e ly i n te r m s o n ly o f
T h u s, any r e f e r e n c e t o an o b je c t
m u st b e a b l e t o b e r e s o l v e d i n t o a s e t o f r e l a t i o n s h i p s d e f i n e d o n p r im ­
i t i v e s ; t h e r e l a t i o n s h i p s a n d t h e d e s c r i p t i o n s o f t h e p r i m i t i v e s m u st h a v e
b een s to r e d .
O n ly K S 's d e s c r i b i n g t h e s t r u c t u r e s o f c o m p o s ite s a r e s t o r e d
i n a n y th in g l i k e t h e i r o r i g i n a l f o rm : KS's d e s c r i b i n g p r i m i t i v e s a r e d e ­
c o m p o sed , e n d t h e d e s c r i p t i o n s a r e s t o r e d i n t h e p r i m i t i v e l i s t ,
and l a t e r
i n t h e s y s te m p r i m i t i v e l i s t .
A c h ie v in g t h i s
•
i s r e l a t i v e l y s im p le . ^W e d o i t a s f o llo w s :
T h e s y s te m c o m p o s ite l i s t
is
i n i t i a l i s e d w i t h t h e m a in
sy s te m
g o a l a s i t s o n l y m em ber. '
•
When f g o a l i s in v o k e d , t h e a c t i o n f i e l d o f t h e XS t h a t h a s b e e n
p a ssed to i t , is in s p e c te d .
I f t h e name t h a t a p p e a r s t h e r e , d o e s
n o t a p p e a r i n t h e s y s te m c o m p o s ite l i s t , th e n t h e KS i s ig n o r e d ,,
an d f g o a l t e r m i n a t e s .
THE DESIGN OF THE PROTOTYPE SYSTEM.
.
67
t lis t,
I f t h e name d o c s , h o w e v e r, a p p e a r d
th e
p a ra m e te r
o f t h e name i s
e x a m in e d ;
p r im itiv e , i t is
i n s p e c t e d a n d t h e de sc
p r im itiv e l i s t .
If
c o m p o sed ,
and
e n te re d in
appear
th e
it
d e s c rib e s
if
1
a
m em bers o f t h e i
t h e s y s te m c o m p o s ite l i s t ,
th e re .
( O b j e c ts may a p p e a r i r
o n ly a s t h e o p e r a n d s o f r e l a t i o n s h i p s , w h ic h e x p l a i n s why
c o n d itio n
s ta c k i s
o p e ra n d
se arch e d ,
sta c k i s
but its
s e a r c h e d .)
th e
T h e c o n d i t i o n o p e ra n d
m em bers a r e n o t p o p p e d ; t h i s
is
so
t h a t a t t h i s s t a g e , a l l t h r e e s t a c k s g e n e r a t e d b y t h e KS may b e
e n t e r e d i n t o t h e s y s te m s t r u c t u r e l i s t a s a r e c o r d u n d e r t h e name
o f th e o b je c t.
W h e th e r t h e o b j e c t d e s c r i b e d i s a p r i m i t i v e o r
a c o m p o s ite , i t s name i s now rem o ved fro m t h e s y s te m c o m p o s ite
•
T h is p r o c e s s c o n t i n u e s u n t i l t h e s y s te m c o m p o s ite l i s t i s e m p ty ,
when t h e g o a l h a s b e e n
A lth o u g h
no
il l y f a c to re d .
c h e c k s h a v e b e e n im p le m e n te d i n t h e p r o t o t y p e ,
it
is
easy
en o u g h t o s e e t h a t i f t h e e n t i r e k n o w ^ d g e b a s e f o r t h e b lo c k f g o a l h a s
been
c o n s u lte d ,
and
can n o t be fa c to r e d .
th e
s y s te m
c o m p o s ite l i s t
i s u n c h an g e d , th e g o a l
T h i s w o u ld b e c a u s e d e i t h e r b y a s p e l l i n g e r r o r , o r
b y an i n s u f f i c i e n t k n o w le d g e b a s e .
T h is a lg o r i t h m h a r d l y
fu rth e r,
re q u ir e s
f u r t h e r com m ent; we w i l l n o t d i s c u s s
an d w i l l move t o a d i s c u s s i o n o f how t h e
t e s t s m e rg e s.
THE DESIGN OF THE PROTOTYPE SYSTEM.
it
sy s te m g e n e r a t e s an d
How the system generates and tests merges.
The " g e n e r a t e a n d t e s t "
te c h n iq u e i s
a l s o w e l l known i n
th e A r t i f i c i a l
I n t e l l i g e n c e l i t e r a t u r e , a n d we h a v e u s e d a v r y much s i m p l i f i e d v e r s i o n
b y l i m i t i n g t h e s c o p e o f b o th m e rg e g e n e r a t i o n a n d m erg e t e s t i n g .
M erges
a r e g e n e r a t e d o n l y u p t o t h e p r i m i t i v e l e v e l ; a f t e r t h a t , t h e m e rg e s r e ­
s u l t fro m f o r w a r d c h a i n i n g th r o u g h t h e s t r u c t u r a l d e s c r i p t i o n .
T h e " m erg e g e n e r a t e " d u t i e s l i e s o l e l y w i t h t h e k n o w le d g e b a s e : K S 's i n ­
fo rm t h e g tm e r b l o c k w h ic h r e g i o n s s h o u ld b e t e s t e d w i t h a v ie w t o m e rg ­
ing-.
T h e p r o c e s s im p le m e n te d i n t h e p r e s e n t s y s te m i s e x tr e m e ly s im p le ,
a n d g o e s as f o l l o w s :
•
The
b lo c k
e v a lu a te s
K S 's an d r e c o r d s t h o s e c a n d i d a t e s rec o m ­
m ended f o r m e rg in g b y K S 's w h ic h e v a l u a t e t o g r e a t e r t h a n 0 ,7 5
tru e
(th is
fig u r e
is
p u re ly a r b it r a r y )
a n d c o n tin u e s t o - d o s o
u n t i l i t h a s e v a l u a t e d a l l K S 's r e l e v a n t t o
in th e p i c t u r e .
it
f o r a l l r e g io n s
T h i s i s u n w ie ld y , g r a n t e d , b u t t h e s y s te m i s a
p ro to ty p e .
’e‘‘
•
When t h e
e v a lu a to r i s
s a tis fie d ,
a l l m erg e c a n d i d a t e s a r e
c o v e re d , and th e v a lu e s t h a t w o rld r e s u l t f o r te x t u r e ,
g rey
re ­
av erag e
l e v e l a n d c e n t r e o f g r a v i t y f o r t h e C a n d id a te r e g i o n , a r e
found.
( T h e s e p a r a m e t e r s a r e n o t c la im e d t o b o t h e
best
fo r
t e s t i n g a m e rg e .)
A t t h i s s t a g e , we t e s t o n ly t h a t t h e new a v e r a g e g r e y l e v e l an d
te x tu r e
a re th e
sam e a s t h o s e o f t h e m erg e c a n d i d a t e s ,
THE DESIGN OP THE PROTOTYPE SYSTEM.
but i t
i s i n te n d e d t h a t the* p r i m i t i v e l i s t
s o p h is tic a te d ,
be c o n s u lte d t o
g o a l - d i r e c t e d m e rg in g .
a llo w
more
I f th e r e s u l ts a re s a t ­
i s f a c t o r y , t h e c a n d i d a t e s re m a in i n t h e m erg e l i s t .
•
T h e m e rg e l i s t
is
th e n
passed
to
t h e n e ig h b o u r - c o tie
m e rg in g
s e c t i o n , an d t h e p i c t u r e r e p r e s e n t a t i o n i s u p d a te d a c c o r d i n g l y .
A lth o u g h t h i s
se g m e n t o f t h e p r o t o t y p e i s s t i l l e x tr e m e ly p r i m i t i v e , • i t
/
i s a se g m e n t t h a t w i l l a llo w g r e a t e x p a n s io n a s i t
o f a c t i v e k n o w le d g e ’ c o u ld b e s e e n t o b e p r o d u c t i v e .
i s h e re t h a t th e u se
T he i n t e n t i o n w o u ld
b e t o add t o th e g e n e r a te s t a g e 's l i s t o f c a n d id a te r e g io n s th o s e re g io n s
t h a t s t r u c t u r e i n d i c a t e s may w e l l b e m e rg e d t o fo rm a m is s in g p r i m i t i v e .
The t e s t s t a g e w o u ld b e i n s t r u c t e d t o b e l e n i e n t i n i t s t e s t i n g o f t h e s e
c a n d i d a t e s , a n d s h o u ld a l s o p a s s o u t i n f o r m a ti o n i n d i c a t i n g how " g o o d " a
m e rg e i s p e rf o rm e d u n d e r t h e s e c o n d i t i o n s .
The suggestions d e s c r i b e d a b o v e w o u ld le n d g r e a t p o w er t o t h e s y s te m , b u t
a r e f a r t o o s o p h i s t i c a t e d t o h a v e b e e n im p le m e n te d i n t h e p r e s e n t p r o t o ­
ty p e .
We w i l l d i s c u s s n e x t how t h e p r o t o t y p e f i n d s a g o a l i n a p i c t u r e .
A c t iv e k n o w le d g e w as m e n tio n e d i n t h e i n t r o d u c t i o n a s a c o n c e p t i n ­
tr o d u c e d a s f a r a s I c a n s e e , 'b y F r e u d e r i n h i s 1976 t h e s i s .
i s t h a t i f o n e h a s fo und p a r t o f an o b je c t,
w h e r e t o lo o k f o r t h e r e s t o f i t .
THE DESIGN OF THE PROTOTYPE SYSTEM.
T he id e a
o n e h a s a g o o d i d e a ’ off
F i n d in g t h e g o a l.
The f i r s t p o i n t t o n o t e h e r e i s
th a t th is
s e c ti o n o f th e p ro to ty p e h a s
n o t y e t b e e n im p le m e n te d 6 ; t h u s , t h e 'a l g o r i t h m d e s c r i b e d h a s n e i t h e r b e e n
t e s t e d , n o r d e e p ly e x a m in e d , a n d p o t e n t i a l im p le m e n ta tio n
d if f ic u ltie s
h a v e n o t y e t b eco m e a p p a r e n t .
T he p r o c e s s t h a t w o u ld h a v e b e e n u s e d i s a n o t h e r o n e f a m i l i a r i n t h e A r ­
t i f i c i a l In te llig e n c e
lite ra tu re ,
t h a t o f f o rw a rd c h a in in g .-
T h is w o u ld
h a v e b e e n d i r e c t e d to w a rd f i n d i n g t h e m a in s y s te m g o a l , g iv e n t h e l o c a t i o n
o f th e p r im it iv e s .
F o r th e p ro ce ss to s t a r t ,
m a tc h e s o u s t a l r e a d y h a v e
b e e n f o u n d f o r a l l m em bers o f t h e s y s te m p r i m i t i v e l i s t .
We in t e n d t h e
p r o c e s s t o f o llo w t h e s e s t e p s :
•
T he s t r u c t u r a l p r i m i t i v e
lis t
has
been
in itia lis e d
w ith
th e
n am es o f t h e a b s o l u t e p r i m i t i v e s ; i t now c o n t a i n s ( b e c a u s e t h e s e
p r i m i t i v e s h a v e b e e n m a tc h e d ) , a g a i n s t e a c h name a l i s t o f t h e
r e g i o n n u m b e rs w h ic h m a tc h t h a t p r i m i t i v e .
T he g o a l f i n d e r m u st
now f i n d i n t h e s y s te m s t r u c t u r e l i s t a d e s c r i p t i o n o f a n o b j e c t ,
T h is w as a s a r e s u l t o f p io b le m s w ith tim e , an d t h e
fa c t th a t
th e
' ‘t e c h n i q u e o f r u n n in g b lo c k s a s se g m e n ts t o f i t th em i n t o t h e 32K s p a c e '
a llo w e d , w as f a i l i n g :
t h e se g m e n ts th e m s e lv e s w ere g e t t i n g t o o b i g .
Any im p l e m e n ta tio n o f t h e a lg o r i th m d e s c r i b e d w o u ld h a v e e n d e d u p i n
t h e g tm e r s e g m e n t,
re-a rra n g e m e n t
w h ic h
s im p ly
o f t h e p r o t o t y p e 's
THE DESIGN OF THE PROTOTYPE SYSTEM..
had
no
sp ace
le ft,
and
a
m a jo r
s t r u c t u r e w o u ld h a v e b e e n n e c e s -
w h ic h i s d e s c r i b e d o n ly i n te r m s o f r e l a t i o n s h i p s b e tw ee n mem­
b e rs o f th e
s tr u c t u r a l p r im itiv e
lis t.
I f such a d e s c r ip tio n
c a n n o t b e f o u n d , t h e g o a l c a n n o t b e f o u n d , an d t h e s y s te m f a i l s .
•
A d a ta s t r u c tu r e
is
s e t u p , w h ic h c o n t a i n s a n o r d e r i n g o f t h e
s y s te m p r i m i t i v e s r e q u i r e d f o r t h e o b j e c t t o b e f o u n d , r e c a l l i n g
t h s t LEG1 an d LEG2, f o r e x a m p le , r e f e r t o d i s t i n c t l e g s , a n d m u st
t h e r e f o r e b e m a tch e d t o d i s t i n c t r e g i o n s ,
•
"I’h e r u l e
is
e v a lu a te d f o r t h i s
t h e name o f t h e o b j e c t i s
o r d e r in g :
if
it
is
s u c c e s s f u l,
d e m o ted t o t h e s t r u c t u r a l p r i m i t i v e
l i s t , i f i t i s n o t t h e r e a l r e a d y , a n d t h e m a tc h in g s e t i s e n t e r e d
a g a in s t i t .
•
Tf i t i s u n s u c c e s s f u l, n o th in g i s done.
A new o r d e r i n g o f t h e p r i m i t i v e s i s now t r i e d , a n d t h e p r o c e s s
c o n t i n u e s e i t h e r u n t i l t h e s y s te m
u n a b le
to
f i n d s KS which is i n
fa ils
as a re s u lt
of
b e in g
terms only of members o f th e
s y s te m p r i m i t i v e l i s t , o r u n t i l t h e s t r u c t u r e l i s t i s e m p ty , i n
w h ic h c a s e t h e sy s te m h a s f o u n d t h e g o a l .
I t c a n b e s e e n t h a t t h i s i s a s im p le t e c h n i q u e , an d t h a t i t i s i n a f a i r l y
p r i m i t i v e f o r a , b u t t h a t i t W ill a llo w o f e x p a n s io n , an d c o u ld w e l l becom e
p o w e r f u l»
T he f i r s t n e c e s s i t y f o r t h e te c h n i q u e t o b ecom e t r u l y p o w e r f u l w o u ld b e
t h e p e r f o r m in g o f some form o f s e n s i t i v i t y a n a l y s i s o n t h e KS, t o d e t e r ­
m in e w h ic h p r i m i t i v e s w ere a b s o l u t e l y n e c e s s a r y f o r t h e KS t o e v a l u a t e
as tr u e .
A t t h i s s t a g e , n o t b e in g a b l e t o m a tc h a l l a b s o l u t e p r i m i t i v e s
i s f a t a l ; w i t h a m ore s o p h i s t i c a t e d s 'y ste m , t h e c o n c e p t o f a c t i v e know­
THE DESIGN OF.THE PROTOTYPE SYSTEM.
le d g e may b e em p lo y ed t o a llo w m o re l e n i e n t c l a s s i f i c a t i o n o f r e g i o n s t h a t
a p p ea r in th e r i g h t p la c e f o r p r im it iv e s , b u t have n o t been so c l a s s i f i e d .
T h u s , f o r a p o w e r f u l s y s te m , we n e e d some fo rm o f tw o -w a y i n f o r m a ti o n flo w
b e tw e e n
g tm e r
and t h i s
s e g m e n t,
a llo w in g e a c h b lo c k t o
r e c o n s id e r I t s
d e c i s i o n s i n t h e l i g h t o f new e v id e n c e u n c o v e r e d b y t h e o t h e r b l o c k .
T h is
f r e e r f lo w o f i n f o r m a t i o n w i l l r e q u i r e a g r e a t d e a l o f w o rk t o im p le m e n t,
b u t s h o u ld g e n e r a t e a s y s te m o f e n o rm o u s p o w er a n d c o n s i d e r a b l e p o t e n t i a l .
CONCLUSION.
I n t h e a b o v e d i s c u s s i o n s , we h a v e shown t h e d e s ig n o f a p r o t o t y p e s y s te m
i n te n d e d t o show t h e f e a s i b i l i t y a n d d e s i r a b i l i t y o f b u i l d i n g t h e s y s te m
d e s c r i b e d i n t h e c h a p t e r o n t h e new p a ra d ig m .
Many o f t h e s e d e s ig n d e ­
c i s i o n s w e r e ta k e n d u r i n g t h e b u i l d i n g o f t h e s y s te m , a lth o u g h t h e m a jo r
s t r u c t u r e o f t h e d e s i g n h a d b e e n s k e tc h e d o u t b e f o r e h a n d .
d e s ig n
has
p r o d u c e d a s y s te m w h ic h w o rk s t o
As i t i s , t h e
th e e x te n t o f s tim u la tin g
c u r i o s i t y , a n d i n my v ie w , o n e w h ic h h a s r a i s e d m o re q u e s t i o n s th a n i t
h a s an sw ered .
Some o f t h e m ore i n t e r e s t i n g o n e s a r e n o te d b e lo w :
•
How c a n i n e x a c t k n o w le d g e b e h a n d le d i n a s y s te -
♦
How c a n a c t i v e k n o w le d g e
■'
We h a v e d e s c r i b e d f o rw a rd c h a i n i n g t o t h e g o a l .
o f t h i s k in d ?
d e p lo y e d r e a l l y e f f e c t i v e l y ?
Can t h i s b e made
t o w o rk w e l l w i t h an in f o r m a ti o n i n t e r c h a n g e b e tw e e n t h i s b l o c k ,
an d t h e g e n e r a t e and t e s t m erg e b lo c k ?
THE DESIGN OF THE PROTOTYPE SYSTEM.
75
•
J u s t whaL l e v e l o f e x p l i c i t k n o w le d g e i s n e c e s s a r y i n t h i s s y s ­
tem ?
Do we n e e d t o d e f i n e different "knowledge languages" & ith
d i f f e r e n t c a p a c i t i e s f o r d i f f e r e n t b l o c k s , t o an y g r e a t e r e x t e n t
t h a n we h a v e d o n e a l r e a d y ?
•
W ill e l a r g e s e t o f m e rg in g K S 's m anage e l l c l a s s e s o f p i c t u r e s ,
o r o u s t s e t s o f m e rg e K S 's w h ic h a r e p o t e n t i a l l y c o n t r a d i c t o r y , .
b e a s s o c i a te d w ith d i f f e r e n t c la s s e s o f p ic tu r e s ?
•
W h a te v e r t h e a n sw e r t o t h e p r e v i o u s q u e s t i o n , how c a n t h e s e KS’ s
be d is c o v e re d ?
T h e s e a r e among t h e m a jo r i s s u e s r a i s e d b y t h e d e s ig n , an d p r o v id e g ro u n d
f o r e x te n s iv e re s e a rc h .
THE DESIGN OF THE PROTOTYPE SYSTEM.
THE IMPLEMENTATION OF THE PROTOTYPE.
As we h a v e b r i e f l y m e n tio n e d t h e s e g m e n t a ti o n o f t h e c o d e , t h e la n g u a g e
o f im p l e m e n ta tio n a n d t h e s y s te m o n w h ic h t h e s y s te m was im p le m e n te d , we
w i l l n o t d is c u s s th e s e is s u e s
a g a in .
We w i l l d i s c u s s
t h e way i n w h ic h
t h e p r o t o t y p e w as im p le m e n te d i n te rm s o f t h e d a t a s t r u c t u r e s , t h e e r r o r
h a n d l i n g , t h e te c h n i q u e s u s e d t o c o d e t h e s y s te m , a n d t h e c o d e g e n e r a t e d .
THE DATA STRU CTU RES.
As we h a v e s t a t e d a b o v e , t h e s y s te m d a t a s t r u c t u r e s w ere c o d e d a s p r o c e ­
d u r e s , w ith t h e d a t a in t h e o r i g i n a l im p l e m e n ta tio n b e in g l o c a l t o t h e
p r o c e d u r e ; o p e r a t o r s a n d d a t a w ere t h e n p a s s e d i n
t u r e s a s p a ra m e te rs f o r t h e s e p r o c e d u re s .
an d o u t o f t h e s t r u c ­
We h a v e s a i d a b o v e t h a t tw o
c a t e g o r i e s o f d a t a a r e r e c o g n i s e d : s y s te m d a t a , w h ic h a r e r e q u i r e d e i t h e r
b y m o re th a n o n e KAB, o r b y o n e KAB o n m o re th a n o n e o c c a s io n , an d s im p le
d a t a , w h ic h e x i s t l o c a l t o a b lo c k .
S y stem d a t a h a d t o b e p l a c e d i n
a common b lo c k d e c l a r e d i n
th e lo a d e r ,
a s w e l l a s i n t h e b l o c k , a s t h e H P-1000 when i t lo a d e d a se g m e n t, o v e r ­
w r o t e t h e p r e v io u s s e g m e n t 's l o c a l d a t a s p a c e , w h ic h w as n o t r e s t o r e d when
th e p r e v i o u s se g m e n t was r e l o a d e d ; o n ly d a t a common w i t h t h e l o a d e r w ere
s a fe .
T h is
s itu a tio n
is
not
id e a l
fro m
a
s o f tw a r e
e n g in e e r i n g
p o i n t - o f - v i e w , an d le'd t o an a m ount o f d e b u g g in g t h a t w o u ld o t h e r w is e h a v e
THE IMPLEMENTATION OF THE PROTOTYPE.
been u n n e c e s s a r y ,
c au se d by e r r o r s
su c h a s m is s p e lt, common b lo c k n am es,
and i n a p p r o p r i a t e a c c e s s t o s t r u c t u r e s .
T he c o m p ile r u s e d s t o r e d common
b lo c k s c o n t i g u o u s l y , w ith no run-tim er1b o u n d
s u b s c r ip te d v a r ia b le s .
a b o u t 75% o f t h e
ch e c k in g on
ac c e sse s to
T h is c r e a t e d an a p p a l l i n g o v e rh e a d i n d e b u g g in g :
e rro rs
f o u n d , w o u ld h a v e b e e n f o u n d an d r e j e c t e d b y a
l a n g u a g e t h a t c h e c k e d b o u n d s o n s u b s c r i p t e d v a r i a b l e s a t r u n tim e .
S im p le d a t a d i d n o t i n g e n e r a l p r e s e n t s u c h p r o b le m s .
T hese d a ta ( fo r
e x a m p le t h e s t a c k s g e n e r a t e d b y a r u l e when i t w as b e in g e v a l u a t e d ) w ere
m o re t r a c t a b l e , l a r g e l y a s n o m o d i f i c a t i o n s h a d t o b e made t o p r o c e d u r e s
u s i n g s im p le d a t a , when t h e c o d e h a d t o b e se g m e n te d .
D a ta s t r u c t u r e s w e re c o d e d t o
r e t u r n , d e p e n d in g o n t h e d a t a s t r u c t u r e ,
p a r a m e t e r s o t h e r th a n t h e d a t a r e q u e s t e d .
ro u tin e s
th a t
c o m p ris e
t h e s y s te m a r e
I n a p p e n d ix I , w h e re t h e s u b ­
i n d i v i d u a l l y d e s c r i b e d , we h av e
d o c u m e n te d t h e m e an in g o f r e t u r n e d p a r a m e t e r s .
ERROR HANDLING.
v.
T he e r r o r h a n d l i n g p r o v id e d i n t h e im p l e m e n ta tio n i s
at th is
q u ire d to
a p a r a m e t e r w h ic h i n d i c a t e s w h e th e r i t
c essfu l
r e tu r n
o r w h e th e r i t
l e v e l o f im p l e m e n ta tio n .
s im p le , b u t i t . h a s
p ro v ed e f f e c tiv e
has f a ile d ;
in
g e n e ra l,
s y s te m m u st f a i l , a lth o u g h t h e r e a r e e x c e p t i o n s .
THE IMPLEMENTATION OF THE PROTOTYPE.
E ach r o u t i n e i s r e ­
has
been
i f a ro u tin e f a i l s ,
suc­
th e
When a r c - u tin c d e t e c t s a c o n d i t i o n a s a r e s u l t o f w h ich i t w i l l
fa il,
it
w i l l w r i t e a m e ssa g e t o t h e o u t p u t d e v ic e d e s c r i b i n g t h e f a i l u r e an d i t s
s o u r c e ( t h e num ber o f t h e c u r r e n t K S ). I t w i l l th e n s e t i t s
r a m e te r t o
z ero ,
s ta tu s p a ­
a n d w i l l t e r m i n a t e , b y g o in g t o t h e s t a t e m e n t n u m b ered
" 7 0 " (7 0 w as c h o s e n f o r n o p a r t i c u l a r r e a s o n ) .
H en c e, i n a n y r o u t i n e ,
" g o to 7 0" m eans t h a t t h a t r o u t i n e s h o u ld f a i l an d t e r m i n a t e .
A ro u tin e
t h a t c a l l s a r o u tin e t h a t th e n f a i l s , w i l l i t s e l f f a i l , u n le s s th e e r r o r
is
o n e o f t h e s m a ll c l a s s o f n o n - r a t a l e r r o r s t h a t w i l l b e
T h u s, th e
d e s c rib e d .
f a i l u r e o f a n y o n e r o u t i n e c a u s e s a c h a in o f f a i l u r e s , u n t i l
t h e lo a d e r i t s e l f
f a ils .
I n m o st c a s e s , e r r o r s w i l l b e r e p o r t e d a lo n g
t h e c h a i n , a n d o n e e r r o r may g e n e r a t e a c h a i n o f e r r o r m e s s a g e s .
N o n -f a ta l e r r o r s a re :
*
An a tte m p t t o p o p fro m a n e m p ty c o n d i t i o n s t a c k , i f a t l e a s t o n e
c o n d itio n
has
a lre a d y
been
popped
e v a lu a te d a l l o u r c o n d itio n s ,
( th is
m eans
th a t
we h a v e
a n d m u st move t o t h e n e x t p h a s e
o f e v a l u a t i n g t h e K S ).
♦
An a tte m p t
to
access
r e g i o n 's n e ig h b o u r
n e ig h b o u r s
(th is
in f o r m a t i o n
code,
w ill
ab o u t th e n 't h
w h e n v - th a t
r e tu r n
a
re g io n
has
r e g io n o f a
few er
th a n
n
s t a t u s v a lu e t h a t f o r c e s t h e
c o n d i t i o n v a lu e t o z e r o ) .
T h e s e e r r o r - h a n d l i n g te c h n i q u e s h a v e s o f a r p r o v e d a d e q u a te t o t h e n e e d s
o f d e b u g g in g t h e p r e s e n t s im p le k n o w le d g e b a s e .
THE IMPLEMENTATION OF THE PROTOTYPE.
CODING TECHNIQUES USED FOR BUILDING THE PROTOTYPE SYSTEM.
T he s y s te m w as b u i ? t o n s o u n d
so ftw a re
e n g in e e r i n g
p rin c ip le s .
D a ta
s t r u c t u r e s w ere " h id d e n " fro m h i g h e r - l e v e l 'p r o c e d u r e s , b y u s i n g t h e a p ­
p ro a c h d e s c rib e d .
T he c o d e i t s e l f w as s t r u c t u r e d , a lth o u g h i n some p r o ­
c e d u r e s t h e s t r u c t u r e _ i s m e ssy ,
l a r g e ly a s a r e s u l t o f th e p r a c t i c e o f
p u t t i n g " f i x e s " i n , w ith o u t p e r f o r m in g an y r e v i s i o n s t o r e g a i n s t r u c t u r a l
e le g a n c e .
A l l r o u t i n e s a r e r e q u i r e d t o i n d i c a t e w h e th e r th e y h a v e f a i l e d
o r s u c c e e d e d a n d , a lth o u g h o n e may a rg u e t h a t i t i s p o s s i b l e f o r 6 r o u t i n e
t o b e u n a b le , b y v i r t u e o f i t s
f a i l i n g , to r e p o rt t h a t i t h as f a i l e d , th e
te c h n i q u e w ork s w e l l e n o u g h i n p r a c t i c e .
T he la n g u a g e u s e d was s t a n d a r d FORTRAN 77 w i t h tw o e x c e p t i o n s .
•
An o p t i o n p r o v i d e d
t h e c o m p ile r t o
We h a v e
b y H e w le tt- P a c k a r d t h a t a llo w s o n e t o f o r c e
dem and e x p l i c i t t y p i n g o n a l l v a r i a b l e s , w h ic h
t o a v e r y l a r g e e x t e n t s i m p l i f i e s d e b u g g in g .
•
A r o u tin e s p e c if ic
t o t h e H P-1000 w as u s e d w h ic h r e c o v e r s a p a ­
r a m e te r l i s t w h ic h
i s g iv e n when t h e p ro g ra m i s r u n .
This m e an t
t h a t t h e o u t p u t d e v ic e c o u ld b e s p e c i f i e d a t r u n tim e , b y p a s s i n g
t
t h e p ro g ra m , i n t h e i n v o c a t i o n , t h e num ber o f t h e d e v ic e t o
w h ic h t h e o u t p u t s h o u ld b e d i r e c t e d .
T he DO WHILE s t a t e m e n t ,
a lth o u g h s t a n d a r d t o FORTRAN 7 7 , was n o t u s e d ;
t h e s t r u c t u r e shown b e lo w w as u s e d i n s t e a ’d , l a r g e l y ' a s 1 d i d n o t r e a l i s e
u n t i l t o o l a t e , t h a t FORTRAN 77 p r o v id e d t h i s s ta te m e n t .
• THE IMPLEMENTATION OF THE PROTOTYPE.
,
76
aaa
CONTINUE
I F ( i n v e r s e c o n d i t i o n ) GOTO bbb
( s t a r t o f b lo c k )
)
(
(e n d o f b lo c k )
GOTO a a a
bbb
CONTINUE
L o g i c a l v a r i a b l e s w ere n o t u s e d , a s a r e s u l t o f a c o m p ile r b u g t h a t made
t h e i r b e h a v io u r u n p r e d i c t a b l e .
te n d e d t o b e o f th e
to
INTEGER; t h i s
I n s te a d ,
a l l v a r i a b l e s w h ic h
w ere
in ­
l o g i c a l ty p e w e re d e c l a r e d a s INTEGER*2 ( a s o p p o se d
a s s o c i a te s w ith th e
c la s s o f v a ria b le a
u n iq u e
ty p e ,
m a k in g c h a n g e s t o t h e i r d e c l a r a t i o n s im p le ) an d th e n a s s ig n e d e i t h e r z e r o
( f a l s e ) o r one ( tr u e ) a s n e c e s s a ry .
T h e c o d in g - te c h n i q u e s u s e d w ere f a i r l y s t a n d a r d ; t h e c o d e t h a t was g e n ­
e ra te d i s a w e ..in te r e s ti n g .
THE CODE.
T h e p r o t o t y p e s y s te m was c o d ed a s a l a r g e s e t o f s u b r o u t i n e s , w h ic h a r e
i n d i v i d u a l l y d e s c r i b e d i n a p p e n d ix I .
T h is te c h n i q u e a llo w e d se g m e n ta ­
t i o n to be im p le m e n te d e a s i l y when i t was r e q u i r e d ,
fro m w h ic h s i m p l e r su b s y s te m s c o u ld b e c lo n e d .
THE IMPLEMENTATION. OF THE PROTOTYPE.
an d le d t o
a sy s te m
U sin g t-licsc t e c l i n j q u c s , somt' 4300 H n e s w ero w r i t i e i j ,
to
iraplBmcnL th e
p r o t o t y p e d e s c r i b e d a b o v e , w i t h t h e e x c e p tio n o f t h e s e c t i o n o f g tm e r t h a t
c h a in s f o rw a rd th r o u g h t h e s y s te m s t r u c t u r e l i s t t o f i n d t h e m ain sy s te m
g o a l.
T h is c o d e i s
k i t h ' c o m p ile r
p re s e n te d ,
re p o rts
an d t h e
in
its
se g m e n te d f o rm , i n
lo a d m ap.
It
a p p e n d ix I I I ,
c a n b e s e e n w hen c e r t a i n
s u b r o u t i n e s a r e i n s p e c t e d ( " s t r l i s V , t h e s y s te m s t r u c t u r e l i s t ,
fo r ex­
a m p le ) , t h a t t h e r e a r e s e c t i o n s o f c o d e t h a t a r e n o t u s e d b y t h e sy s te m
as i t
is
a t p re s e n t.
T h e s e a r e f o r t h e u s e o f t h e s e c t i o n t h a t w as n o t
im p le m e n te d .
T he a l g o r i t h m s u s e d a r e d e s c r i b e d i n "T he D e s ig n o f t h e P r o t o t y p e S y stem "
a n d t h e n e ig h b o u r - c o d e m e rg in g te c h n i q u e s i n " N e ig h b o u r C o d e s" .
im p le m e n ts t h e s e a lg o r i th m s c o r r e c t l y , a s f a r a s I c a n s e e .
The code
The p r o to ty p e
h a s b e e n t e s t e d s u c h t h a t t h e p r e s e n t v e r s i o n h a s p ro d u c e d n o e r r o r s when
r e q u i r e d t o d e a l w i t h K S 's o f d i f f e r e n t ty p e s i n d i f f e r e n t b l o c k s . T e s t i n g
d o e s n o t a llo w o n e t o
c e rtify
a s y s te m e r r o r - f r e e , b u t a l l
e rro rs
en­
c o u n te r e d h a v e b e e n d e a l t - w i t h , an d I h a v e s e e n n o e r r o r s fro m t h e p r e s e n t
v e r s i o n i n a num ber o f t e s t s .
CONCLUSION.
T he im p l e m e n ta tio n o f t h e p r o t o t y p e h a s b e e n d e s c r i b e d .
We h a v e d i s c u s s e d
t h e i s s u e s a n d te c h n i q u e s i n v o lv e d , r a t h e r th a n t h e d e t a i l s o f t h e c o d e .
T he s y s te m , i t i s c la im e d , w o rk s , an d r e s u l t s w i l l b e g iv e n a n d d i s c u s s e d
i n t h e f o llo w in g c h a p t e r .
THE IMPLEMENTATION OF THE PROTOTYPE.
I n t h i s c h a p t e r , r e s u l t s o b ta in e d fro m w o rk i n t h i s p a ra d ig m w i l l b e d e ­
s c rib e d .
We w i l l a tte m p t t o show t h a t th e y e n c o u r a g e t h e u n d e r t a k i n g o f
f u r t h e r w o rk i n
t h i s p a r a d ig m .
T h e r e s u l t s a r e o f tw o t y p e s ; t h o s e o b ­
t a i n e d w i t h i n t h e p a ra d ig m , b u t n o t a s a r e s u l t o f t h e u s e o f t h e s y s te m
d e s c r ib e d a b o v e , and th e r e s u l t s o b ta in e d by
p ro to ty p in g
th is
s y s te m .
T h e f o rm e r a r e m o re d i r e c t l y e n c o u r a g in g .
RESULTS OBTAINED WITHIN THE PARADIGM.
We h a v e d i s c u s s e d a b o v e t h e w o rk o n t h e p a r t i c l e - s i z e d i s t r i b u t i o n a n a ­
ly s e r .
I t i s n o t im m e d ia te ly o b v io u s t h a t t h i s
i s w ith in th e scope o f
t h e p a ra d ig m o f f e r e d , b u t i t c a n b e s e e n t h a t t h e p i c t u r e ' i s b e in g h a n d le d
i n te r m s o f r e g i o n s ,
a n d t h a t t h e p r o c e s s i s k n o w le d g e - g u id e d , a lth o u g h
t h e k n o w le d g e u s e d i s n o t e x p l i c i t .
T h is w o rk h a s g e n e r a te d some p ro m ­
is in g re s u lts .
T h e p h o to g r a p h s p r o v id e d a s f i g u r e 11 show t h e q u a l i t y o f p i c t u r e ta k e n
o n D e e l k r a a l g o ld m in e , w h ile d a t a w e re b e in g c o l l e c t e d f o r t h i s p r o j e c t .
T h e s e p i c t u r e s a r e h a r d l y c o m p r e h e n s ib le , b u t show g o ld o r e o n a c o n v e y o r
b e lt,
The o r e i s
b e in g c o n v ey e d t o
a m ill,
an d h a s n o t b e en
c ru s h e d .
T he p ro b le m p o s e d b y t h e l a r g e q u a n t i t y o f a d s o rb e d f i n e p a r t i c l e s c an
b e se e n to be c o n s id e r a b le .
RESULTS.
1 -Si
.
81
•
I t c a n a l s o b e s e e n t h a t tli c p r o to ty p e o f th e p a r t i c l e - s i z e d i s t r i b u t i o n
a n a l y s e r (PSDA) h a n d le s t h e p ro b le m f a i r l y w e l l , a s i t h a s se g m e n te d t h e
p i c t u r e , d e s p i t e i t s q u a l i t y , i n t o r e g io n s
s e n s ib ly .
The
PSDA h a s
been
an d h a s m erged t h o s e r e g i o n s
s e e n t o h a n d le many s u c h p i c t u r e s ;
th e
te c h n i q u e s i t u s e s a r e c l o s e l y li n k e d t o t h e i d e a s d e s c r i b e d a b o v e , w i t h
t h e e x c e p tio n t h a t
th e y
a re
s t r u c t u r e o f t h e s y s te m .
in
g e n eral
m ore
d e e p ly
em bedded
in
th e
T h u s , t h e PSDA f i n d s t h e p a r t i c l e s i n t h e p i c ­
tu r e by:
1.
S e g m e n tin g t h e p i c t u r e i n t o r e g i o n s .
2.
M e rg in g t h e s e r e g i o n s a c c o r d in g t o t h e r u l e s d e v is e d .
3.
G o in g o v e r t h e m e rg e d p i c t u r e
le d t o
t o u n d o t h o s e m e rg e s w h ic h h a v e
r e g io n s o f a su s p e c t s h a p e ,
( e .g .
h o u r g la s s - s h a p e d
re ­
g io n s a r e n o t e n c o u r a g e d )
T h is t e c h n iq u e
is
r e d o l e n t o f t h e i d e a s t h a t m o t i v a t e t h e p a ra d ig m p r e ­
s e n t e d , o f a k n o w le d g e - g u id e d p r o c e s s , a n d o f a c t i v e k n o w le d g e .
T hese
id e a s h a v e b e e n b a s i c t o t h e w o rk d o n e b o th b y t h i s a u th o r a n d b y B e r g e r ,
who h a s o b t a i n e d t h e s e r e c e n t r e s u l t s ’ w i t h t h e PSDA.
T he p o i n t h e r e i s
t h a t t h e s e i d e a s c a n l e a d t o u s e f u l w o rk in g s y s te m s .
P
T he PSDA was d e v e lo p e d u s i n g r e g i o n s a s t h e f e a t u r e f i r s t e x t r a c t e d , a f t e r
i t b e c a m e 'a p p a r e n t t h a t t h e y w o u ld b e t h e s i m p l e s t f e a t u r e t o w o rk w it h .
(T h e p r o c e s s d e s c r i b e d a b o v e w o u ld b e f a r m o re cum bersom e i f
to
be
e x tra c te d .)
T hus,
th is
w o rk
can
e d g e s w ere
b e c la im e d t o b e w i t h i n t h e
p a ra d ig m p r o p o s e d on t h i s C o u n t a s w e l l , an d i t s p r e s e n t a p p a r e n t s u c c e s s
can b e u se d t o s tr e n g th e n th e j u s t i f i c a t i o n
RESULTS.
f o r t h e " p arad ig m .
9 : The a b o v e phoL ograpl- is show th e r e s u l t s oi! th e PCDA w o rk ,
The lo p p h o lo i’n ip h i s ' o t th e p i c t u r e ta k e n on th e m in e : th e b o tto m
K1G
th e tiC G in i'n iiitio n a r r i v e d a t Ijy th e PSDA.
RESULTS OBTAINED BY PROTOTYPING THE SYSTEM.
I c a n n o t c la im y e t t h a t t h e p r o t o t y p e
*->■ I s ,
and g e n e r a te s , t e s t s ,
g io n s , b u t th e
la tte r
is
a n a ly s in g s c e n e s ;
a c tiv ity
f a c to r s
i s n o t y e t g u id e d to w a rd t h e p r i m i t i v e s
r e q u i r e d , o r to w a rd f i n d i n g an o b j e c t i n t h e p i c t u r e .
o u t,.
it
c o n fir m s a n d im p le m e n ts m e rg e s b e tw e e n r e ­
We w i l l show h e r e
rom t h e s y s te m a s i t ' e x p a n d s a g o a l , an d m e rg e s r e g i o n s , a lth o u g h
t h e m e rg in g i s n o t g o a l - d i r e c t e d .
R e fer to
f ig u r e
1 0 , w h ic h show s t h e k n o w le d g e b a s e o n w h ic h t h e sy s te m
o p e r a t e s i n t h i s e x a m p le .
I t c o n s i s t s o f 12 K S 's , t h e 1 2 't b of w h ic h is
n o t u s e d , a s t h e KAB f i n i s h h a s n o t b e e n im p le m e n te d , w ith t h e r e s u l t t h a t
t h e s y s te m w i l l c y c l e e n d l e s s l y th r o u g h t h e r u l e b a s e , w hen i t s t a s k s h a v e
b e e n p e r f o r m e d , a s o p p o se d t o te r m i n a t i n g c l e a n l y .
T h e K S 's w i l l b e i n ­
t e r p r e t e d b e lo w .
T h e r e a r e o n ly tw o K S 's w i t h " s t a r t " i n t h e i r SBC f i e l d , an d t h e i r m eaning
h a s b e e n g iv e n a b o v e , a s t h e KS’ s u se d i n t h e s t a r t b lo c k d o n o t v a r y .
T h e K S 's w i t h f g o a l s i n t h e i r SBC f i e l d ‘ fo rm t h e d e s c r i p t i o n o f a n e l e ­
p h a n t,
a lth o u g h a n a r b i t r a r y
h a v in g f o u r l e g s ,
N o tic e
o n e . T he
tw o e y e s , tw o e a r s ,
e le p h a n t
has
been
d e s c rib e d
a p o t e n t i a l s o u rc e o f c o n fu s io n h e re .
The ta g s
i n t h e SBC
f i e l d s f o r f g o a l and g tm e r a r e f g o a l s an d g tm e r g , f o r e n t i r e l y
to r ic a l
rea so n s,
c o m p le x itie s .
RESULTS.
as
and a tr u n k , w ith no s p a t i a l t e -
h is ­
an d c h a n g in g them w o u ld h a v e in v o lv e d u n n e c e s s a r y
la tio n s h ip s p e c if ie d
b e tw e e n t h e s e p r i m i t i v e s .
T he d e s c r i p t i o n
may
be
t r i v i a l , an d i n a c c u r a t e , b u t i t i s s u f f i c i e n t t o i l l u s t r a t e t h e p r o c e s s .
We h a v e d e s c r i b e d a l e g a s c o n s i s t i n g o f a t h i g h o n to p o f a k n e e o n to p
o f a s h i n , an d " e y e " , " e a r " , " t r u n k " , " t h i g h ” , " s h i n " a n d " k n e e " a r e a l l
p r i m i t i v e s , w h o se d e s c r i p t i o n a lth o u g h o p e n t o a rg u m e n t, i s s u f f i c i e n t l y
c o m p re h e n s ib le i n t h e KS la n g u a g e t o n e e d n o f u t h e r d i s c u s s i o n .
i s o n l y o n e r u l e w i t h g tro e rg i n
its
SBC f i e l d ,
T h ere
w h ic h e x p r e s s e s t h e
in ­
t e n t i o n th a t- r e g i o n s w i t h n e ig h b o u r s w i t h t h e sam e g r e y - l e v e l , s h o u ld b e
m e rg e d .to th e m .
.At t h i s s t a g e o f o u r d e m o n s t r a t i o n , we m u st com pose a p i c t u r e a n d r e n d e r
i t i t i t o t h e l l e p r e s e n t a t i o n f o r t h e s y s te m t o uses, a s we a r e n o t y e t u s i n g
t.he s o f tw a r e e s c r i b e d t o B e r g e r i n t h e t e x t a b o v e , f o r c o m p u tin g n e ig h b o u r
i:c d e s , b e c a u s e i t
ivt^aft.ito ta k e o n e .
ta tio n
is
e a s i e r t o com pose a p i c t u r e t o t h e s y s t e m 's l i k i n g
T he p i c t u r e we h a v e u s e d i s
it:g e n e ra te s
in fig u re
11.
shown w ith t h e r e p r e s e n ­
T he o u t p u t o f th e s y s te m f o r t h i s
I'ljictu i-e, -g iv e n t h i s k n o w le d g e b a s e , i s show n i n f i g u r e 1 2 .
We h a v e a n -
\■itT'tatsdx t h e o u t p u t a s b e in g m o re u s e f u l i n t h a t fo rm , t h a n i f
it
w e re
d e s c r iv n d . V
What i s i W r t h n o t i n g i s :
*
<
Vhe sys-tern h a s f a c t o r e d t h e g o a l s u c o . j s f u l l y , an d h a s a f u l l
- :/ id e s e i r ip t io n o f i t i n te r m s b o th o f p r i m i t i v e s a n d o f s t r u c t u r e .
•
The
s y s te m
has
g e n e ra te d ,
mes.'jWf; b e tw e e n r e g i o n s ,
te s te d ,
and th e
c o n firm e d
r e s u lts
c o r r e c tly r e f le c te d in th e re p r e s e n ta tio n .
RESULTS.
of
a n d im p le m e n te d
th e se
m e rg e s
a re
Thus t h e p r o t o t y p e i s d o in g w h at we h a v e c la im e d i t c a n d o .
p e r h a p s c l e a t i s why t h i s
I t i s i m p o r ta n t b e c a u s e i t
W hat i s n o t
i s im p o r ta n t.
i s £. c l e a r i n d i c a t i o n t h a t a p r o t o t y p e o f t h i s
n a t u r e c a n b e s u c c e s s f u l l y b u i l t , an d c a n b e e x p e c te d t o w ork w ith s im p le
p ic tu re s .
It
is
ro o st u n f o r t u n a t e t h a t t h e p r o b le m s t h a t p r e v e n t e d t h e
im p l e m e n ta tio n o f t h e b lo c k t h a t w ou ld h a v e m e rg e d f o r w a r d t o
t h e m ain
s y s te m g o a l , o c c u r r e d when th e y d i d , a s t h e s u c c e s s o f t h i s b lo c k w o u ld
h a v e made t h e p r o t o t y p e a m ore c o n v in c in g a rg u m e n t f o r f u r t h e r w o rk to w a rd
b u i l d i n g a m o re s o p h i s t i c a t e d v e r s i o n o f t h e s y s te m d e s c r i b e d ; a s i t i s ,
th e p ro to ty p e i s
o p e n t o t h e a rg u m e n t s o o f t e n s u c c e s s f u l a g a i n s t w o rk
i n t h e a r t i f i c i a l - i n t e l l i g e n c e f i e l d , " S u r e , i t lo o k s g o o d now , b u t w h at
h a p p e n s when i t e n c o u n t e r s t h e r e a l p r o b le m s ? " .
T h is i s u n f o r t u n a t e , a s t h e a rg u m e n t i s s t r o n g , a n d a s we h a v e s a i d a b o v e ,
th e re
d o e x i s t p r o b le m s f a r h a r d e r th a n t h e o n e s s o lv e d h e r e , many o f
w h ic h w i l l n o t r e s p o n d w e l l t o b e in g s i d e s t e p p e d .
c ritic is m
I m u st a n sw e r
th is
f r a n k l y , b y s a y in g t h a t I am n o t c e r t a i n t h a t t h e s e p ro b le m s
may b e s o lv e d i n t h a t t h e s y s te m i n t h e fo rm d e s c r i b e d , w i l l a n a ly s e r e a l
s c e n e s , I ro u st e x p r e s s m o re f a i t h i n t h e p a ra d ig m d e s c r i b e d (T h e e v id e n c e
in i t s
favour i s
s t r o n g e r . ) , w h ic h n o t o n ly h a s r e a s o n a b l e t e n e t s , b u t
w h ic h c a n a l s o b o a s t a w o rk in g system ( t h e PSDA) d e s ig n e d w i t h i n i t .
T he p r o t o t y p e c a n b e s t b e d e s c r i b e d a s p a r t o f an u n f i n i s h e d lo n g - te r m ;>
e x p e r im e n t: i t h a s w o rk e d t o t h e e x t e n t t h a t i t h a s b e e n im p le m e n te d , an d
t h i s i n d i c a t e s t h a t f u r t h e r w o rk o n i t w o u ld b e f r u i t f u l .
fo rm ,
it
s tr o n g
c a n b e m ade t o w o rk on s im p le p i c t u r e s ,
a rg u m e n t
fo r
it
I f in th e f u l l
w ould r e p r e s e n t
h e r e , w ith a v ie w t o b u i l d i n g a n e v e n m ore p o w e r f u l s y s te m .
RESULTS.
a
a l a r g e r r e s e a r c h e f f o r t i n t o t h e p ro b le m s r a i s e d
I am u n a b le
'
85
-k; w h at I c a n s a y
say h e re t
;hy i t
rfo rm e d t o
s h o u ld n ’t ,
L n a t th e work
and o f f e r s h o p e
C O N C L U S IO N .
.We h a v e show n r e s u l t s
fro m w o rk o n t h e p a r t i c l e - s i z e
d is trib u tio n
l y s e r , an d fro m t h e p r o t o t y p i n g o f t h e s y s te m d i s c u s s e d .
ana­
T he o u t p u t o f
t h e s y s te m f o r a g iv e n k n o w le d g e b a s e w as sh o w n , a n d t h e p r o t o t y p e sy s te m
c o u ld b e s e e n t o
f a c to r a g o a l,
p le m e n t m e rg e s .
T hese r e s u l t s do
p a ra d ig m
is ju s t if i e d ,
p r e s e n te d a b o v e.
and t o
g e n e ra te ,
c o n f ir m
th a t
t e s t , c o n f ir m a n d im ­
fu rth e r
b u t do n o t y e t f u l l y v a l i d a t e
w ork
in
th is
th e 's y stem d e s ig n
T h e r e i s , h o w e v e r, h o p e f o r t h e d e s i g n , a s a lth o u g h i t
has n o t been v a lid a t e d ,
t h e p r o t o t y p e s u g g e s ts t h a t i t may b e .
S tro n g
a rg u m e n ts may s t i l l b e r a i s e d a g a i n s t t h e d e s ig n a n d t h e p r o t o t y p e , b u t
f u r t h e r w o rk a lo n g t h e s e l i n e s seem s t o b e j u s t i f i e d .
RESULTS.
OOM GOA LS
‘ ( ( ( ( ( ( ( ( H A S i L E G l) W lS<LEE2>)A HAS<LEG?))A HA5(LEG4: )A HAS
< E Y E !))A HAS<EYE2>)A HAS<EAR1>)A W S (E A R 2 ))A H A 5 (T R U f« l> )A 1ELEFHT<COI1Pi«
0D3«FGDALS
‘ (m < K N E E l,T H lE m > a N (S H IN l,K lE E V )A « L E G < C O T )'
(l04'F£O A L5
»(SHA PE(RCUI0> SlZE<UShALL> )A 'E Y E tPR ll1> ‘
QOS'START
M M DB5
«tRlE(OU»>*ClCTXT<FGOALS>*
•SihG<CENTRE,fil>*M EREE<CENTRE,Rl)*
O aV FG O A L S •((({ S H A P E iR O lH D > TEX1U K (!.OW >)A E F L E C < U K > )A
)A S 1 2 E < L A R £ E ')A 'E S )R < P R ]IV *
: a U<I1EDIUM>
FIG. 10,1 : Tlie K n o w led g e b ase used f o r t he ex am p le .
V L P r r LEC<La-' ft. Tf*
CO V 'FG Q W .5
J lff'F G O A L S
* ( S H t f t< L * T ) S IZ E iM E O lU m M 'T K lG H tP R lIV *
S I Z £ ( S m u L > ) S 1K E £ ( P R m “
0 1 J* F S D ftL S
« (S H A P E < l» T > S iZ E C to D I U li! S '6 H lH iP R I M > »
D 1 2 * F IH i5 H
«!R tJE<D ’J I1)*C H C T X T <FlH IS H >‘
FIG. 10,2 : The K n o w led g e b a s e u s e d ' o r th e e x a m p le .
Region
1
2
3
4
f ig .
11
Code
2 1-12 4 3
1 3 4 1-1
14 Z
1 23
Are qreu lerel
0,0
0,0
63,0
63,0
The picture used in the example,
with its representation.
1kp
r tto v i^
itjstewv |
4k
wvm
a o c o l-
AGEKGOAD
SYSTEM PEEX1IRES
CORRECT S P E L L IN G ,< - 6 CHAPS
C H CTX KFG OA LS)
FSO A LS
RNUM
6
MtETMW
S H lN <PR IM >
FIG.12.1: Hie o u tp u t o f th e system fo r the g iv e n knowledge base and th e gi.
p ic tu re .
A.
H W f
IS
PIG 12,2: The o u tp u t o f the system f o r Che g iv en knowledge base and th e given
p ic tu re .
&
:<C E N T R E.R 1>
E P » £ < C E H T R E ,R V
1.000
4HB5GE
2
<
3
YIO W .5; T h e o u t p u t o f th e s y s te m f o r th e g iv e n k n o w le d g e b a s e a n d t h e g iv e n
p ic tu r e .
fm
4
RNUH
RtMl
FIG,42.4= T he o u t p u t o f th e s y s te m f o r th e g iv e n k n o w le d g e b a s e a n d t h e g iv e n
p ic tu re .
6
STATE
4
P R fiU S T FOLLOWS
EYE ROUIO
EAR R O U ® M EO IU U D U DK
TR L M : L «T 'D K
liED IUtlLO U
TH IG H V Y
KNEE ROUND
th e . p n w t W
r
'
STRUCTURE F O L LO JS
A TER STA tK
I ANDSTACK
IN T H IS ORDER
I
CANDSTfCK
RTORSTACK
A )A )A )A )A )A )A )A )((((((((
ANDSTACK
HAS HAS HAS
HAS HAS
CANDSTACK
TRUNKIEAR2 EAR1 EY E2 E Y E I
ATORSTACK
HAS
HAS
LEG4 LEG}
HAS
HAS
L E S2
LEG!
ANDSTACK
ON
ON
”
l ” m T H lM tM B
NCTIffi FOLLOWS
j-'
-1 -1 2 4 ? B 0 0 H 0
% _ _ _ _ _ 1 } 4 1- 1 0 0 00 0 0 # 0
14 2 0 0 B 0 0 0 0 0 D
A_____1 2 3 0 0 0 0 G 0 0 0 0
0 0 00 0 0 9 0 0 0 0 0
o o
^
th u is ■ lk e
4
0 e Cll
1 /5
6 0 1
D0 }
0 ob
U
o o o o a o o o o o o o lo
0 0
0 C
0 0 0 0 0 9 0 9 C 0 0 0 [G
00 0 0 0 0 0 0 0 0 0 0 p
a o oo o o o o o b a g o o p
9 0 G0 0 G G 0 0 0 0 0 0 0 i)
NCXREF
_________________ —
IXf' 5 6 7 B M O
rE ? L ^ — —
a
to
j
-
4
I .
KJUH
FIG 42.5; The o u t p u t o f th e s y s te m f o r th e g iv e n k n o w le d g e b a s e a n d t h e g iv e n
C O N C L U SIO N
A new p a ra d ig m t o
a llo w g e n e r a l s c e n e a n a l y s i s h a s b e e n p r e s e n t e d .
A
s y s te m h a s b e e n d e s ig n e d w i t h i n t h i s p a ra d ig m , a n d p r o t o t y p e d i n a l i m i t e d
fo rm , t o show t h a t t h e p a ra d ig m d o e s in d e e d s i m p l i f y t h i s p r o b le m .
d e s i g n . a n d im p l e m e n ta tio n o f t h e p r o t o t y p e h a v e b e e n d i s c u s s e d ,
o u tp u t o f th e p ro to ty p e ,
p re s e n te d .
T he
and th e
o p e r a t i n g o n a g iv e n k n o w le d g e b a s e , h a s b e e n
S u p p o r t fro m o t h e r w o rk f o r t h e p a ra d ig m h a s b e e n p r e s e n t e d .
We h a v e s a i d a b o v e t h a t t h i s w o rk b y i t s n a t u r e c o u ld n o t y e t d e m o n s tr a te
t h e v a l i d i t y o f t h e p a r a d ig m , o r o f t h e s y s te m d e s ig n e d w i t h i n i t ,
but
m u st b e l i m i t e d t o sh o w in g t h a t t h e r e h a v e b e e n n o g r o s s e r r o r s i n t h e i r
c o n s t r u c t i o n , a n d t h a t t h e y a r e f e a s i b l e a n d v a l i d g o a l s t o w o rk to w a rd .
I b e l i e v e t h a t t h i s h a s b e e n d o n e , b y t h e p r e s e n t a t i o n b o th o f t h e s u c c e s s
of
th e
PSBA w ork., w h ic h h a s b e e n show n t o h a v e b e e n d o n e w i t h i n t h i s
p a ra d ig m , an d b y t-/..e r e s u l t s o f t h e w o rk o n t h e p r o t o t y p e , w h ic h h a v e b e e n
e n c o u r a g in g , a s we h a v e show n t h a t t h e r e p r e s e n t a t i o n u s e d i s t r a c t a b l e ,
t h a t a p r o t o t y p e c a n b e b u i l t t h a t may h a n d le p i c t u r e s u s i n g t h i s r e p r e ­
s e n t a t i o n ', a n d t h a t t h i s p r o t o t y p e may f a c t o r t h e g o a l t h a t i t
has been
p r e s e n t e d w i t h , i n t o a fo rm o f t h i s r e p r e s e n t a t i o n .
T he
u n f o rtu n a te
asp ect
of
th is
w o rk i s
its
n e ce ssary s u p e r f ic ia lity ;
v i s i o n i s a h a r d p r o b le m , b u t i t i s t r y i n g t o b e i n t h e p o s i t i o n
a d o p t, w h ere o n e c a n s a y o f t h i s
w o rk o n l y t h a t i t
p r e s e n ts
I m u st
a lin e
of
a p p ro a c h b y w h ic h t h e p r o b le m may p o s s i b l y b e s o l v e d , a n d show s t h a t t h i s
a p p ro a c h h a s n o g l a r i n g e r r o r s , a n d t h a t s y s te m s c a n b e b u i l t u s i n g
t h a t w o rk u n d e r c e r t a i n c i r c u m s t a n c e s .
CONCLUSION
it,
REFEREN CES.
Armco
A u to m e tr ic s ,
M anual
fo r
B a r r , A. a n d F e ig e n b a w a , E .A .
a
P a r tic le
S iz e D is tr ib u ti o n A n a ly se r,
"T he H andbook o f A r t i f i c i a l I n t e l l i g e n c e " ,
W illia m K aufm an i n c . ,1 9 6 1 .
B e rg er,
S F -,
B .S c .( E le c .
•‘'F i r c i c l e
E n g .)
S iz e
d is s e r ta tio n
D is trib u tio n
s u b m i tte d
A n a ly s is " ,
to
u n p u b lis h e d
U n iv e rs ity
of
th e
t f i t w a t e r s r a n d , J o h a n n e s b u r g , 1984.
B e rg er, S .F .,
( p r i v a t e c o n v e r s a t i o n s ) , 1 9 8 4 ( a ) , an d 1985.
B rach m an , R . J . , e t a l , "W hat a r e E x p e r t S y s te m s ? " ,
i n H a y e s - R o th ,P .
et
a l ( e d s ) , " B u ild i n g E x p e r t S y s te m s " , A d d is o n -W e s le y , 1983.
B u c h an a n ,
R .J .
and
S h o r tliff e ,
E .H .,
" R u le - b a s e d
E x p ert
S y s te m s " ,
A d d iso n -W e sle y , 1 9 8 4 .
C h a r n ia k ,
E .C .
"A
Common
L an g u a g e -G o n ip re h e n sio n
R e p r e s e n ta tio n
In fo r m a tio n " ,
fo r
A rtific ia l
P r o b le m - s o lv in g
and
I n te llig e n c e
v o l . 16
I n te llig e n c e
v o l . 15
p p 2 2 5 -2 5 5 , 1981.
D a v is, R .,
" R e a s o n in g
p p l7 9 - 2 2 2 , 1 9 8 0 .
REFERENCES.
about
C o n tro l" ,
A rtific ia l
, D a v i s , R , , " C o n te n t R e f e r e n c e : R e a s o n in g a b o u t R u l e s " , A r t i f i c i a l I n t e l ­
l i g e n c e v o l . 15 p p 2 2 J - 2 4 0 , 1980 ( a ) .
D u b o is , D. an d P r a d e , H . , " F u z z y S e ts a n d S y s te m s: T h e o ry a n d A p p lic a ­
t i o n s " , A ca d em ic, 1 9 8 0 .
F o r s y t h , D .A ., " P a r t i c l e S i z e D i s t r i b u t i o n A n a l y s i s " , u n p u b lis h e d B .S c .
d i s s e r t a t i o n s u b m i tte d t o U n i v e r s i t y o f t h e W iw a te rs ra n d , J o h a n n e s b u r g ,
F r e u d e r , B .C .,
"A C o m p u ter S y stem f o r V i s u a l
R e c o g n i tio n
U s in g
A c t iv e
K n o w led g e ", A .I .- T R - 3 4 5 , M . I . T . , 1 9 7 6 .
H a y e s - R o th , F . e t a l ,
"An O v erv ie w o f E x p e r t S y s te m s " i n H a y e s - R o th , F .
e t a l , ( e d s ) " B u ild i n g E x p e r t S y s te m s" , A d d is o n - W e s le y , 1 9 8 3 .
L e n a t,
D .B .,
"T h e N a t u r e
of
H e u r is tic s " ,
A rtific ia l
In te llig e n c e
19
p p l8 9 - 2 4 9 , 1962
M a rr , D .
, " R e p r e s e n t in g an d C o m p u tin g V is u a l I n f o r m a tio n " i n
P .H . an d Brow n, R .H . ( e d s ) " A r t i f i c i a l I n t e l l i g e n c e : a n M .I .T .
W in s to n ,
P e rsp ec ­
t i v e " , M . I .T . p r e s s , 1 9 7 9 .
M a n d e lb r o t, B . , "The F r a c t a l G eo m etry o f N a t u r e " , W illia m F re e m a n , 1 9 6 2 .
N a z i f , A .H ., an d L e v i n e , M .D .,
"L o w -L ev e l Im age S e g m e n ta t io n :
an E x p ert
S y s te m " , I . E . E . E . t r a n s . P a t t e r n A n a l y s is a n d M a ch in e I n t e l l i g e n c e , v o l .
6 n o . 5 ,o p 5 5 5 - 5 7 6 , S e p te m b e r , 1 9 8 4 .
REFERENCES.
N iem an n , H. a t a l ,
"A K now ledge B ased S y stem f o r A n a l y s is o f G a te d B lo o d
P o o l S t u d i e s " , I . E . E . E . t r a n s . P a t t e r n A n a l y s is an d M ach in e I n t e l l i g e n c e ,
v o l . 7 n o . 3 , p p 2 4 6 - 2 5 9 , May, 1985.
,
( p r i v a t e c o n v e r s a t i o n ) , 1984.
SPIDER M a n u al, MITSUI c o r p . , 1985.
R E A D IN G S .
Many
good
in tro d u c tio n s
to
im ag e p r o c e s s i n g a n d s c e n e a n a l y s i s e x i s t :
am o n g st t h e b e t t e r a r e b o o k s b y R o s e n f e ld a n d K ak , P r a t t , G o n z a le s an d
W in tz , B a l l a r d an d Brow n, a n d Duda an d H a r t .
A g o o d i n t r o d u c t i o n t o A r­
t i f i c i a l I n t e l l i g e n c e i s " A r t i f i c i a l I n t e l l i g e n c e " , b y P .H . W in s to n , a n d
"T he H andbook o f A r t i f i c i a l I n t e l l i g e n c e " c i t e d a b o v e .
READINGS.
APPENDICES
APPENDICES
APPENDIX A. THE SUBROUTINES INVOLVED
The s u b r o u tin e s w i l l b e p re s e n te d one by o n e , w ith a l i s t o f th e r o u tin e s
c a l l e d b y t h e s u b r o u t i n e , an d an y n e c e s s a r y comments o n t h e s u b r o u t i n e .
T he v a r i a b l e s h a v e b e e n e x t e n s i v e l y a n d c a r e f u l l y d e c l a r e d i n
each ro u ­
t i n e ; we w i l l n o t d o t h i s h e r e .
Program RINT
Calls these subroutines.
•
RMPAK
•
CONMAT
CONHEAD
Comments. RINT i s t h e m a in lo a d e r p r o g ra m : i t c o l l e c t s t h e im ag e r e p ­
r e s e n ta tio n
fro m d i s k ,
a n d th e n s c h e d u le s t h e p ro g ra m s e g m e n ts , a s d e ­
s c rib e d in th e te x t .
Appendix A. THE SUBROUTINES INVOLVED
93
Program INITS
Calls these subroutines
COMPUST
•
PRML1ST
STRLIST
Comments.
IN ITS
is
a
se g m e n t,
w h ic h i s why i t
m u st b e d e f i n e d a s
p r o g ra m , t h a t i n i t i a l i s e s t h e sy s te m d a t a s t r u c t u r e s t h a t r e q u i r e i t .
Program START
Calls these subroutines
•
CLFIND
•
CSSETUP
Appendix A. THE SUBROUTINES INVOLVED
ASETUP
•
APSETUP
•
AHEAD
SEGRT
C o m m e n ts . START i s t h e se g m e n t t h a t i s
b lo c k .
START m u st
e v a lu a te
KS’ s ,
r e f e r r e d t o above as th e s t a r t
a n d h e n c e c o n t a i n s m o st o f t h e KS
e v a l u a t i o n a n d a c t i o n c o d e , w h ic h i s p r e s e n t i n a l l KS e v a l u a t i n g r o u ­
tin e s .
T h is s e c t i o n o f s t a r t d o e s n o t h i n g b u t c a l l t h e a p p r o p r i a t e r o u -
Prcgram FGOAL
C a lls t h e s e s u b r o u t i n e s
»
L1STPG0ALS
•
COMPLIST
Comments. FGOAL i s t h e m ain p ro g ra m f o r f a c t o r i n g t h e m ain s y s te m g o a l.
FGOAL c a l l s
th e n e c e ssa ry r o u tin e s , b u t does l i t t l e
Appendix A. THE SUBROUTINES INVOLVED
'
i t s e l f b u t p e rf o rm
95
a n e le m e n ta r y c h e c k on w h e th e r t h e m ain sy s te m g o a l i s
b a s e , w h ic h w o rk s l i k e
th is :
in
t h e k n o w led g e
I f t h e p r e s e n t KS num ber i s l e s s th a n t h e
num ber o f t h e p r e v io u s KS s e e n b y FGOAL, an d t h e m ain s y s te m g o a l i s s t i l l
i n t h e s y s te m c o m p o s ite l i s t , th e n we h a v e c y c le d
th r o u g h t h e k n o w le d g e
b a s e w i t h o u t s e e i n g t h e m a in s y s te m g o a l , a n d i t i s n o t t h e r e .
Program GTMER
Calls t h e s e subroutines
•
6TI2F
‘
•
BRHLIST
\
s m is t
C o m m e n ts . GTMER i s
g fp M .
t h e se g m e n t t h a t g e n e r a t e s a n d t e s t s m erg es f o r r e -
T h is p ro g ra m a g a i n s im p ly c a l l s t h e a p p r o p r i a t e
s v ; r o u t i n e s d o t h e w o rk .
Appendix A, THE SUBROUTINES INVOLVED
ro u tin e s :
th e
Subroutine AGET
Calls these subroutines
•
COMPLIST
Comments. AGET im p le m e n ts t h e a c t i o n GET, w h ic h r e c o v e r s t h e m ain sy ste m
g o a l fro m t h e o p e r a t o r , an d p u t s i t i n t o t h e s y s te m c o m p o s ite l i s t .
Subroutine AHEAD
Calls these subroutines
•
MRECORD
•
CHCTXT
Comments. AHEAD c a l l s t h e a p p r o p r i a t e r o u t i n e t o im p lem e n t t h e a c t i o n
d e s c r i b e d b y t h e e c t i o n name i n t h e KS a c t i o n f i e l d .
Appendix A. THE SUBROUTINES INVOLVED
Subroutine ANDSTACK
C a lls t h e s e s u b r o u t i n e s (NONE CALLED)
C o m m e n ts . ANDSTACK im p lem e n ts
th e
o p e ra n d
s ta c k
( c o n ta in s
c o n d itio n
n a m e s, n o t c o n d i t i o n o p e ra n d n a m e s).
Subroutine APSETUP
Calls these subroutines (NONE CALLED)
Comments. APSETUP r e c o v e r s t h e p a r a m e t e r l i s t t h a t i s a s s o c i a t e d w ith
th e a c tio n in th e a c tio n f i e l d
o f t h e c u r r e n t KS, an d r e t u r n s
a r r a y o f p a r a m e t e r n am es.
Subroutine ASETUP
Calls these subroutines
Appendix A. THE SUBROUTINES INVOLVED
it
as an
C o m m e n ts . ASETUP r e c o v e r s t h e a c t i o n f i e l d from th e KS,
s t r i p p e d o f t h e n u m b e r, SBC an d c o n d i t i o n f i e l d s .
Subroutine ATORSTACK
Calls these subroutines (NONE CALLED)
Cobiments. ATORSTACK im p le m e n ts t h e o p e r a t o r s t a c k d e s c r i b e d i n t h e m ain
I'-s
Subroutine BACKE
Calls these subroutines
Comments. Same as "ROCKE" (q .v .), b u t f o r m em b ersh ip o f t h e b a c k g ro u n d
Appendix A. THE SUBROUTINES INVOLVED
S u b r o u tin e BTS
Calls these subroutines (NONE CALLED)
C o m m e n ts . B ts m akes a lo u d n o i s e a t t h e o p e r a t o r t e r m i n a l i f t h e sy ste m
Subroutine CANDSTACK
Calls these subroutines (NONE CALLED)
Comments. CANDSTACK im p le m e n ts t h e c o n d i t i o n o p e ra n d s t a c k .
Subroutine CHCTXT
Calls these subroutines (NONE CALLED)
Comments. CHCTXT c h a n g e s t h e SBC t o t h e v a lu e p a s s e d i n , im p le m e n tin g
t h e CHCTXT a c t i o n .
Appendix A. THE SUBROUTINES INVOLVED
Subroutine CLEAN
Calls these subroutines (NONE CALLED)
Comments.
CLEAN c l e a n s
up
a n e ig h b o u r c o d e a s d e s c r i b e d i n
a b o v e , b y re m o v in g m u l t i p l e i n s t a n c e s a n d e n d - ir o u n d i n s t a n c e s .
Subroutine CLFIND
Calls these subroutines
Comments. CLilND f i n d s t h e c o n d i t i o n l i s t i n t h e KS, and
BTS i s c a l l e d i f a n e r r o r i s e n c o u n e r e d .
Subroutine CMERGE
Calls these subroutines
•
INSTNO
Appendix A, THE SUBROUTINES INVOLVED
SEARCHI
Comments.
CMERGE i s u s e d t o
rem ove s e t s o f i n t e r n a l r e g io n s fro m th e
c o d e s t h a t a r e a b o u t t o b e m e rg e d .
(S e e t e x t f o r d e t a i l s . )
Subroutine COMPUST
Calls these subroutines (NONE CALLED)
Comments. COMPLIST i s t h e s y s te m c o m p o s ite l i s t , m e n tio n e d i n t h e t e x t
above.
Names may '
in th e l i s t .
T.
M e d t o t h i s l i s t o n ly i f th e y d o n o t m a tc h a n y name
i'M< • • ■b e s e a r c h e d , an d r e t u r n s a p a r a m e t e r t h a t i n ­
d i c a t e s w h e th e r i u
o f th e
lis t
is
- i n s a m a tch t o t h e p a t t e r n s u b m i tte d .
T h e le n g t h
r e t u r n e d a s a p a r a m e t e r , a s when i t i s e m p ty , t h e sy s te m
g o a l h a s b een fa c to re d .
Subroutine CONHEAD
Calls these subroutines
Appendix A. TOE SUBROUTINES INVOLVED
102
Comments. CONHEAD loads the appropriate block, and passes
S u b ro u tin e CONMAT
C alls th e s e s u b r o u tin e s (NONE CALLED)
Com m ents. CONMAT t e s t s t h e SBC f i e l d
c u r r e n t SBC.
S u b ro u tin e CSSETUP
C alls th e s e su b ro u tin e s
•
ATORSTACK
«
ANDSTACK
CANDSTACK
•
REVERSE
Appendix A. THE SUBROUTINES INVOLVED
of
th e
c u rre n t
Comments.
CSSETUP sets up the condition stacks in the fashion described
S u b ro u tin e CVARFCL
C alls th e s e s u b ro u tin e s
•
NUMEVAL
•
RNUMEVAL
•
QUALEVAL
C o m m e n ts .
CVARFCL, a s i t s
name i m p l i e s ,
is
a c lo n e o f CVARFIX, w h ic h
a llo w s a n y r e g i o n r e f e r e n c e t o a p p e a r t o g e t h e r w i t h CENTRE i n a c o n d i t i o n
p a ra m e te r l i s t .
CVARFIX w i l l n o t p r o d u c e u sa b l-> r e s u l t s w ith t h i s
o f p a ra m e te r l i s t ,
a s i t w as I n te n d e d t o b e u s e d w i t h l i s t s
ty p e
c o n s is tin g
o f a r e g i o n r e f e r e n c e a n d an y o t h e r ty p e o f p a r a m e t e r ( f o r ROCK, f o r e x ­
a m p le ) .
A maximum o f tw o r e g i o n r e f e r e n c e s may a p p e a r i n a n y p a r a m e t e r
l i s t ; i f tw o a r e u s e d , o n e m u st b e CENTRE.
Appendix A. THE SUBROUTINES INVOLVED
Subroutine CVARFIX
Calls th e s e su b ro u tin e s
•
NUMEVAL
•
RNUMEVAL
•
QUALEVAL
C om m ents.
CVARFIX
T he v a r i a b l e s
fix e s
th e
a re p a ssed in
c o n d itio n
as s tr in g s ,
in a s e t o f a r r a y s , one f o r each ty p e .
v a r ia b le s as d e s c r ib e d above.
a n d CVARFIX r e t u r n s
t h e v a lu e s
T h u s i f we w ish e d t o f i x t h e p a ­
r a m e te r s t r i n g CENTRE, Q+, 1 00 , t h e a r r a y p a s s e d ( u v a r ) i n w o u ld c o n t a i n
a s i t s f i r s t e le m e n t t h e s t r i n g "CENTRE” , a n d s o o n .
The f ix e d p a ra m e te rs
w o u ld b e r e t u r n e d a s f o l l o w s : T h e f i r s t p a r a m e t e r w i l l b e i n r e g v a r ( l )
( a r e g i o n t y p e ) , t h e se c o n d i n q u a l v s r ( 2 )
i n n u m v a r(3 )
( a n u m b e r).
(a q u a lif ie r ) ,
T h u s t y p i n g i s p e rf o rm e d .
an d t h e t h i r d
T he r o u t i n e c a n n o t
h a n d le t h e s y s te m p r i m i t i v e ty p e y e t . (S e e t h e comment o n CVARFCL, b e f o r e
u s in g t h i s r o u t i n e .)
Appendix A. THE SUBROUTINES INVOLVED
S u b ro u tin e EDGE
C alls th e s e su b ro u tin e s
•
CANDSTACK
•
CVAKFIX
C om m ents. EDGE im p le m e n ts t h e e d g e c o n d i t i o n , w h ic h h a s o n e o p e r a n d , a
re g io n r e f e r e n c e ,
an d r e t u r n s t h e v a lu e 1 . i f t h e r e g i o n r e f e r r e d t o i s
a i e d g e, and a 0.
o th e r w i s e .
As w i t h a l l
c o n d i t i o n s t h a t u s e a r e g io n
r e f e r e n c e o p e r a n d , EDGE u s e s v e c t o r o p e ra n d s an d r e t u r n s v e c t o r r e s u l t s .
S u b ro u tin e GTMERCON
C alls th e s e s u b ro u tin e s
•
CLFIND
•
GSSETUP
•
VECREVAL
•
ASETUP
Appendix A. THE SUBROUTINES INVOLVED
il
106
APSETUP
•
MBROK
•
NCFU1
•
HEWSTATS
C o m m e n ts .
GTMERCON g e n e r a t e s ,
b e tw e e n r e g i o n s .
in th e i r
SBC f i e l d
te s ts ,
c o n fir m s
and
T he g e n e r a t e p h a s e o c c u r s f i r s t :
im p le m e n ts m erges
a l l K S 's w i t h g tm e rg
a r e e v a lu a te d f o r a l l r e g io n s in th e p i c t u r e .
c a n d id a te s a r e re c o rd e d on d i s k . :
M erge
T h e t e s t a n d c o n fir m p h a s e o c c u r s n e x t .
A l l recom m ended m e rg e s a r e p a s s e d t o MEROK, w h ic h a c c e p t s o r r e j e c t s th e m ;
if
th e y a r e a c c e p t e d , th e y a r e a u t o m a t i c a l l y c o n fir m e d b y
th e
c o n fir m
p h a s e , w h ic h o c c u r s n e x t .
T h e c o n f ir m p h a s e t h e n c a l l s NCFUL, a n d HEWSTATS t o u p d a te t h e r e p r e s e n ­
t a t i o n o f th e p ic tu r e .
N o te t h a t GTMERCON i s in v o k e d r a t h e r s t r a n g e l y , a s i t s t a r t s i n t h e f i r s t
p h a s e , a n d i s in v o k e d o n c e f o r e v e r y KS f o r e v e r y r e g i o n , l e a v i n g te m p o ­
r a r y r e s u l t s e a c h tim e .
w hen g tr a e r i s
I t t h e n c h a n g e s p h a s e , b u t c a n s t i l l o n ly e x e c u te
lo a d e d b y r i n t ;
in v o k e d when a . g tm e r KS i s
p h ase,
an d t e r m i n a t e s
v a ria b le .
th u s ,
in th e
l a s t tw o p h a s e s , i t i s o n ly
f o u n d , e x e c u te s u n t i l
it
d e c id e s t o
c h an g e
l e a v i n g t h e num ber o f t h e n e x t p h a s e a s a sy s te m
T h is may s o u n d c o m p le x , b u t i t w as i n
f a c t t h e e a s i e s t way I
c o u ld s e e o f g e t t i n g t h e r o u t i n e t o w o rk a s I w a n te d i t t o .
Appendix A. THE SUBROUTINES INVOLVED
I
107
Subroutine INSTNO
Calls th e s e s u b r o u tin e s (NONE CALLED)
C om m ents.
INSTNO c o u n ts t h e num ber o f i n s t a n c e s o f a g iv e n r e g i o n i n
a n o th e r r e g i o n ' s n e ig h b o u r c o d e .
S u b ro u tin e LGBUF
C alls th e s e s u b ro u tin e s (NONE CALLED)
Com m ents. T h is i s
a
H e w le tt- P a c k a r d
ro u tin e
th a t
e n la r g e s
th e
d is k
b u ffe r a re a .
S u b ro u tin e LISTPGOALS
Calls th e s e s u b ro u tin e s
•
CLFIND
•
CSSETUP
Appendix A. THE SUBROUTINES INVOLVED
108
ASETUP
•
NAHFINB
COMPLIST
•
STRLIST
•
PRMLEXP
C o m m e n ts . LISTPGCA'uS t a k e s a KS, s e t s u p t h e a p p r o p r i a t e s t a c k s ( w i th
CSSETUP), r e t r i e v e s t h e name o f t h e o b j e c t b e in g d e s c r i b e d ,
i f t h e name i s
i n t h e s y s te m c o m p o s ite l i s t .
and checks
I f i t i s , th e d e s c r ip tio n
i s p a s s e d t o t h e sy s te m s t r u c t u r e l i s t , a n d th e n t o PRMLEXP.
I f i t i s n 't ,
t h e r o u t i n e i g n o r e s t h e KS, an d t e r m i n a t e s .
S u b ro u tin e MATCHP
Calls th e s e s u b ro u tin e s (NONE CALLED)
Com m ents. MATCHP m a tc h e s a g iv e n s t a c k a g a i n s t t h e s y s te m p r i m i t i v e l i s t
(w h ich c o n t a i n s b o t h a b s o l u t e an d s y s te m p r i m i t i v e s ) t o a s c e r t a i n w h e th e r
t h e KS w hose c o n d i t i o n o p e ra n d s t a c k i t i s , may b e u s e d n e x t f o r fo rw a rd
c h a in in g t o th e g o a l.
I f a l l t h e n am es i n t h e s t a c k , a r e i n t h e p r i m i t i v e
l i s t , t h e KS may b e u s e d .
Appendix A. THE SUBROUTINES INVOLVED
Subroutine MFROK
Calls these subroutines
(NONE CALLED)
Com m ents. MEROK t e s t s m e rg e s , an d r e t u r n s a f l a g t o a c c e p t o r
r e je c t
S u b ro u tin e MRECORD
C alls th e s e s u b ro u tin e s
CVARFCL
Com m ents. MRECORD r e c o r d s d e t a i l s o n d i s k o f a m erg e t h a t h a s b e e n r e ­
commended b y a KS.
T he d e t a i l s a r e r e c o r d e d o n d i s k s im p ly b e c a u s e th e y
w e re i n t h e e a r l i e s t s y s te m b u i l t , an d t h i s h a s n o t b e e n rem oved fro m t h e
S u b ro u tin e NAMELIST
C alls th e s e s u b r o u tin e s (NONE GALLED)
Appendix A. THE SUBROUTINES INVOLVED
110
C o m m e n t s . NAMELIST h o ld s t h e p o i n t e r s f o r s t r l i s t ,
s tr u c tu r e re c o rd s.
an d t h e nam es o f th e
A " n e x t p o i n t e r " f a c i l i t y h a s b e e n p r o v id e d
in th is
S u b ro u tin e NAMFIND
C alls th e s e s u b ro u tin e s (NONE CALLED)
C om m ents. NAMFIND i s a r o u t i n e c lo n e d fro m APSETUP, t h a t r e t r i e v e s from
a KS t h a t d e s c r i b e s
an o b j e c t ,
t h e name o f t h e o b j e c t d e s c r i b e d ,
and
w h e th e r i t i s a p r i m i t i v e o r a c o m p o s ite , b o th o f w h ic h p i e c e s o f i n f o r ­
m a tio n a r e i n t h e a c t i o n f i e l d o f t h e KS.
S u b ro u tin e NCFUL
C alls th e s e su b ro u tin e s
•
CLEAN
•
NCMERGE
C om m ents.
n e ig h b o u r
NCFUL i s
codes,
th e
w h ic h
c a llin g
checks
ro u tin e
in
Appendix A. THE SUBROUTINES INVOLVED
fo r
m e rg in g
ro u tin e s
u s in g
t h e n e ig h b o u r c r o s s r e f e r e n c e t a b l e
111
w h e th e r t h e r e g io n s h a v e b e e n m e rg e d , a n d c l e a n s u p t h e c o d e s b e f o r e th e y
a r e m e rg e d .
S u b ro u tin e NCMERG
C alls th e s e su b ro u tin e s
•
. INSTNO
•
ROTATE
•
CMERGE
C om m ents. NCMERG u s e s CMERGE t o rem o v e i n t e r n a l r e g i o n s , an d l i n k s t h e
n e ig h b o u r c o d e s when t h i s h a s b e e n d o n e .
S u b ro u tin e NEWSTATS
C alls th e s e s u b ro u tin e s (NONE CALLED)
Com m ents. NEWSTATS c a l c u l a t e s t h e a v e r a g e g r e y l e v e l , g r e y l e v e l v a r i ­
a n c e , c e n t r e o f g r a v i t y , a n d nu m b er o f p i x e l s o f a c o m p o s ite r e g i o n fo rm ed
b y a m e rg e .
Appendix A. THE SUBROUTINES INVOLVED
112
Subroutine NNUME
Calls th e s e s u b ro u tin e s
'
CANDSTACK
•
CVARFIX
Com m ents. NNUME i s t h e p r o c e d u r e c o r r e s p o n d i n g t o t h e c o n d i t i o n NNUli,
w h ic h h a s tw o c o n d i t i o n o p e r a n d s , a r e g i o n r e f e r e n c e a n d a n u m b e r.
if
t h e r e g i o n r e f e r r e d t o h a s t h e g iv e n num ber o f n e ig h b o u r s , NNUME a s s i g n s
t h e v e c t o r t h e v a lu e 1 . i n t h e c o r r e s p o n d i n g e le m e n t, w h ic h o th e r w is e i s
a s s ig n e d 0 .
NNUM a t p r e s e n t i s u s e d w i t h o n ly CENTRE a s a l e g a l r e g i o n
re fe re n c e , fo r s im p lic ity .
S u b ro u tin e NUMEVAL
C alls th e s e s u b ro u tin e s (NONE CALLED)
Com m ents. NUMEVAL r e t u ' - s t h e n u m e r i c a l v a lu e o f a s t r i n g o f c h a r a c t e r s
r e p r e s e n t i n g a n u m b e r, f o r e x a m p le , t h e s t r i n g " 1 0 0 " i s e v a l u a t e d t o t h e
num ber 1 0 0 .
Appendix A. THE SUBROUTINES INVOLVED
Subroutine PRMLEXP
C alls th e s e su b ro u tin e s
•
COMPLIST
•
PRHLIST
•
ANDSTAGK
•
CANDSTACK
C o m m e n ts .
PRMLEXP e x p a n d s
s y s te m g o a l .
th e
p rim itiv e
lis t
by
f a c t o r i n g t h e m ain
PRMLEXP i s p a s s e d a K S 's c o n d i t i o n f i e l d ,
t h e name o f t h e
o b j e c t d e s c r i b e d b y t h e KS, an d w h e th e r i t i s a p r i m i t i v e o r a c o m p o s ite .
If it
is
a p r im itiv e ,
its
d e s c rip tio n
i s p la c e d in th e p r im it iv e l i s t ,
t h e o p e r a t o r s b e in g i g n o r e d , an d t h e
c o n d i t i o n o p e ra n d s b e i n g p l a c e d i n
th e f ie l d
F i e l d s e x i s t a t p r e s e n t f o r CONCAV,
fo rc e d by th e c o n d itio n .
SHAPE, TEXTUR, REFLEC a n d SIZ E .
T h e s e d a t a a r e th e n p a s s e d t o t h e p r im ­
itiv e lis t.
I f th e o b je c t i s
a c o m p o s ite , a l l
th e
c o n d itio n
o p e ra n d s
in
its
de­
s c r i p t i o n a r e p a s s e d t o COMPLIST, w h ic h a d d s th em t o t h e s y s te m c o m p o s ite
l i s t , i f th e y a r e n o t t h e r e a l r e a d y (We e m p h a s is e t h a t a n o b j e c t may a p ­
p e a r in a n o th e r o b j e c t 's d e s c r i p t i o n a s a c o n d itio n o p e r a n d .) .
Appendix A. THE SUBROUTINES INVOLVED
T he name o f t h e o b j e c t d e s c r i b e d by t h e KS i s th e n rem oved from th e sy ste m
c o m p o s ite l i s t .
S u b ro u tin e PRMLIST
C a lls t h e s e s u b r o u t i n e s (NONE CALLED)
C o m m e n ts.
PRMLIST im p le m e n ts t h e s y s te m p r i m i t i v e l i s t , w h ic h c o n t a i n s
a r e c o r d o f t h e a b s o l u t e p r i m i t i v e s e n c o u n te r e d when t h e g o a l w as f a c ­
to re d .
T h is s t r u c t u r e may a l s o b e s e a r c h e d f o r m a tc h e s t o a nam e, and
w i l l r e t u r n a l i s t o f t h e r e g i o n n u m b ers t h a t m a tc h t h e p r i m i t i v e .
S u b ro u tin e PUTIT
C alls th e s e s u b ro u tin e s
C o m m e n ts . PUTIT s t o r e s t h e te m p o r a r y r e s u l t s
e v a lu a tio n .
-.n c o u n te r e d
T he a lg o r i t h m i s d e s c r i b e d i n t h e t e x t .
Appendix A. THE SUBROUTINES INVOLVED
d u rin g
ru le
Subroutine QUALEVAL
C alls th e s e s u b r o u tin e s (NONE CALLED)
Com m ents.
QUALEVAL r e t u r n s
a
n u m e ric a l
v a lu e
fo r a q u a lifie r.
T he
p r e s e n t d e f i n e d q u a l i f i e r s a r e : QUITE, VERY, QUITEN, VERYN, DK, Q \ QN-.
Q+ a n d QN- s p e c i f y a r a n g e r a t h e r th a n a v a l u e , a n d a s su c h a r e r e t u r n e d
a s n e g a t i v e n u m b e rs , a n d t h e c o n d i t i o n s u b r o u t i n e s r e c e i v e th em i n t h i s
fo rm .
T he n u m b e rs u s e d a r e n o t im p o r t a n t , b u t a r e c h o se n h e r e t o b e i n
th e ran g e 0 - 1
f o r c o n v e n ie n c e , w i t h n e g a t i v e
n um bers
b e in g
used
to
it.
A
s p e c ify a ran g e .
S u b ro u tin e REVAL
C alls th e s e s u b ro u tin e s
•
VECREVAL
C o m m e n ts . REVAL i s
a h i g h e r l e v e l fo rm o f VECREVAL, an d c a l l s
c a l l t o VECREVAL w i l l r e t u r n t h e v e c t o r v a lu e o f a KS.
A c a l l t o REVAL
w i l l r e t u r n t h e maximum e le m e n t v a l u e o f t h e v e c t o r v a lu e o f a KS.
Appendix A. THE SUBROUTINES INVOLVED
S u b ro u tin e REVERSE
C alls th e se s u b ro u tin e s (HONE CALLED)
Com m ents. REVERSE
n am es a s n e c e s s a r y ( s e e t e x t o n d e co m p o sin g
K S 's ).
S u b ro u tin e RMPAR
C alls th e s e s u b ro u tin e s (NONE CALLED)
C om m ents. T h is i s a H e w le tt- P a c k a r d r
fro m t h e i n v o c a t i o n o f t h e p ro g ra m .
o f t h e m ain p ro g ra m t o b e d i r e c t e d
"RU, H IN T ,2 5 " m eans r u n t h e p r o g ra i
S u b ro u tin e RNUMEVAL
Calls th e s e s u b ro u tin e s
•
NUMEVAL
Appendix A. THE SUBROUTINES INVOLVED
It
d e v ic e s : ;
C o m m e n ts .
RNUMEVAL e v a l u a t e s
an d r e t u r n s
th e r e g io n r e f e r e n c e v e c to r
g e n e r a t e d b y t h e r e g i o n r e f e r e n c e o p e ra n d .
S u b ro u tin e ROCKE
C a lls t h e s e s u b r o u t i n e s
•
CANDSTACK
•
CVARFIX
C o m m e n ts . ROCKE e v a l u a t e s t h e " r o c k n e s s " o f a g iv e n r e g i o n ,
an d c o r r e ­
sp o n d s t o t h e c o n d i t i o n "ROCK", w h ic h h a s tw o o p e r a n d s , a r e g i o n r e f e r e n c e
and a q u a l i f i e
. in t h a t o rd e r.
ROCKE r e t u r n s a v a l u e o f e i t h e r 0 . o r
1- , d e p e n d in g o n W h eth er t h e r e g i o n i n d i c a t e d b e lo n g s t o t h e r o c k c l a s s
to th e d e g re e in d ic a t e d by th e q u a l i f i e r , o r n o t.
T he v a lu e i s r e t u r n e d
a s a v e c t o r , e a c h member o f w h ic h i n d i c a t e s t h e " r o c k n e s s " o f a d i f f e r e n t
c a n d i d a t e f o r t h e g iv e n r e g i o n i d e n t i f i e r ( s e e t e x t ) .
v e c to r
is
re tu rn e d
as
T he l e n g t h o f t h e
w e l l , b e in g - 1 f o r a n i d e n t i f i e r w i t h o n ly o n e
p o s s i b l e m a tch ( l i k e "CENTRE").
Appendix A. THE SUBROUTINES INVOLVED
Subroutine ROTATE
C alls th e s e s u b ro u tin e s
•
ROUND
C o m m e n ts , ROTATE u s e s ROUND t o r o t a t e t h e n e ig h b o u r c o d e u n t i l t h e r e ­
q u i r e d r e g i o n ( n o r m a lly t h e r e g i o n ' s m erg e p a r t n e r ) i s i n t h e f i r s t p o ­
s i t i o n in th e cod e.
S u b ro u tin e ROUND
C alls th e s e s u b ro u tin e s (NONE CALLED)
Com m ents. ROUND r o t a t e s t h e n e ig h b o u r c o d e o n e p l a c e .
S u b ro u tin e RQUEUE
C alls th e s e s u b ro u tin e s (NONE CALLED)
Appendix A. THE SUBROUTINES INVOLVED
Com m ents. RQUEUE im p le m e n ts t h e q u e u e u s e d t o k e e p c o n d i t i o n s i n o r d e r
when t h e y b e e n e v a l u a t e d .
(S e e t e x t f o r d e t a i l s . )
S u b ro u tin e SEARCHI
C alls th e s e s u b ro u tin e s (NONE CALLED)
C om m ents.
SEARCHI
f in d s th e p o s it io n o f t h e n e x t o c c u rre n c e , a f t e r a
g iv e n p o s i t i o n , o f a n i n d i c a t e d r e g i o n i n a n e ig h b o u r c o d e .
S u b ro u tin e SEGLD
C alls th e s e s u b r o u tin e s (NONE CALLED)
Com m ents. SEGLD l o a d s a se g m e n t an d t r a n s f e r s c o n t r o l t o i t :
o t h e r H e w le tt- P a c k a r d r o u t i n e .
Appendix A. THE SUBROUTINES INVOLVED
it
is
an­
■t fr'
/ '
:
S ubi-outine SEGRT
C a lls t h e s e s u b r o u t i n e s (NONE CALLED)
C o m m e n ts'. SEGRT I s a H e w le tt- P a c k a r d r o u t i n e t h a t r e t u r n s c o n t r o l from
t h e se g m e n t t o t h e lo a d e r .
S u b ro u tin e SIMGE
C a lls t h e s e s u b r o u t i n e s
»
CANDSTACK
•
CVARFCL
Com m ents. SIMGE im p le m e n ts t h e c o n d i t i o n SIMG, w h ic h i s r e q u i r e d t o h a v e
a s i t s o p e ra n d s tw o r e g io n r e f e r e n c e s .
A t p r e s e n t one o f th e s e re fe re n c e s
rou st b e t o CENTRE, a s we h a v e n o t m ade p r o v i s i o n f o r c o m p a rin g a l l members
o f a s e t i n a l l p o s s i b l e w ays w ith a l l members o f a n o th e r s e t .
re tu rn e d as th e
A 1.
is
c o n d itio n v a lu e i f th e g re y le v e ls a r e s im ila r , a 0 . i f
th e y a r e n o t.
Appendix A. THE SUBROUTINES INVOLVED
Subroutine SiMTE
C alls th e s e su b ro u tin e s
•
CANDSTACK
•
CVARFCL
Com m ents. SIMTE is s i m i l a r t o SINGE, b u t c o m pares t e x t u r e s ( g r e y - l e v e l
v a r i a n c e ) r a t h e r th a n g r e y l e v e l s .
S u b ro u tin e STRLIST
C alls th e s e s u b ro u tin e s
•
NAMELIST
•
MATCH!
Com m ents.
STRLIST
im p le m e n ts t h e
sy s te m s t r u c t u r e
lis t.
b e e n t e r e d i f t h e i r name d o e s n o t a l r e a d y a p p e a r i n t h e
R e c o rd s may
lis t,
o r d e m o ted t o p r i m i t i v e s , w hen a m a tch l i s t m u st b e s p e c i f i e d .
d e le te d ,
The l i s t
a l s o a llo w s a s e a r c h t o b e made f o r t h e n e x t l e g a l s t r u c t u r a l d e s c r i p t i o n
( i n te rm s p r i m i t i v e s o n l y ) .
T h e r e c o r d s a r e k e p t i n a s im p le a r r a y , th e
Appendix A. THE SUBROUTINES INVOLVED
122
list structure being imposed by NAMELIST, which stores the pointer!- to
S u b ro u tin e TRUEE
C alls th e s e su b ro u tin e s
•
CANDSTACK
C om m ents. TRUEE im p le m e n ts t h e c o n d i t i o n TRUE.
I t a lw a y s p o p s o n e o p ­
e r a n d fro m t h e c o n d i t i o n o p e ra n d s t a c k , a n d a lw a y s r e t u r n s t h e v a l u e 1 .
: t h u s , TRUE m u st a lw a y s h a v e o n e o p e r a n d , c o n v e n t i o n a l l y c a l l e d DUM, b u t
n o t in sp e c te d .
S u b ro u tin e VECREVAL
C alls th e s e su b ro u tin e s
•
RQUEUE
•
ANDSTACK
Appendix A. THE SUBROUTINES INVOLVED
123
%
BACKE
NNDME
•
TRUEB
•
ATORSTACK
C o m m e n ts . VECREVAL e v a l u a t e s a KS, r e t u r n i n g a
e n tir e
KS.
T h e te c h n iq u e u s e d i s
v e c to r
v a lu e
fo r
th e
e x h a u s tiv e ly d e s c r ib e d in th e te x t .
N o te t h a t t h i s s u b r o u t i n e may n o t a lw a y s a p p e a r w ith i t s
f u l l com p lem en t
o f c o n d i t i o n e v a l u a t i o n c a l l s ( c a l l s t o : ROCKE, BACKE, THUEE, EDGE, NNUME,
SIMGE, o r SIHTE) b u t i s
d o c u m en te d h e r e a s n e e d in g them a l l ,
p l i c i t y and g e n e ra lity .
Appendix A. THE SUBROUTINES INVOLVED
fo r
sim ­
APPENDIX B.
M
THE BACKUS NAUR FORM SYNTAX OF THE KS'S.
MOOIM IS OH CR 00041 USING 00012 B-K5 ft=0060
11:1? Mi FRL, IDOl L , I960
00D?
Backus Naur Form Expression Of Rules ,
0005
$006
The rules require the use of the characters < and >jto evoid
iLonCusiori.howveT .i* u it here ( inC >.
00QB
0009
0010
Q011
0012
0015
0014
0015
0016
001?
001B
0019
0620
U021
0022
0023
00 2 4
6625
0026
0027
0028
0029
0030
60?1
0052
0037
0034
0035
0036
003?
003B
003?
0040
0041
<rule)::=(digit)<digit>(digit>*<cl!st>*<slist>
<clist>::«(<operend> (operand))AI(<oper6-.d> (operand))0II(operand))N
(operand)(condition)I(clist>
(condition)::- BMCK((regionno>,(qualifier))I
ROCK<(regionno),(qualifier»l
TRUE<DU«) I
5irKE<<regionno>,<regionno)>I
SMTEUrbgionno) ,(f eoionno>> I
W<obin:me>,(objname>) I
Hr*
jnaBis), (objnasie) >I
r;:aionno),(integer>>
TL.'V '-'tdes»l
SKft£H(»hdes»l
Sl2E{<sides»l
ft£FLEC((rfdes)>l
aHVEX4(cdes»
(Ides’ ' I'K IG K Ih E D lltllL n i!
(iM et.-: i =V T
IPMD
(rfdes)::=HIGHIMEOIUMILOU
{cdes>::,i(quaiifi6r>
(sides>::-BIGILITTLEIMEOiiS1
(regionno>::»R(integer)ICEN1RE
(integer)::=(digit)I(digit>(integer)
(digit>::=0UI21?l4l5l6l7l8l9
(qualifier)::*OUITEIOUlTENIlERYIUERYNIDKim-IQ«
(elist)::=r£SEE((regianno),<merlist))l
CHCTXT<<neucontext>)I
fi£ET(GM.>l
(objnaee)1(objdescriptar)>
(objdescriptor>::'PRIHIC011P
(objn»me)::=(char!ist)
(charlitt):^(char)|(char>(cherlist>
(char>::-AI8ICIDIEIFIGmillJIKILmiNI0IPIBlRISniUIUiyiXIYIZ
(n e rlis t)::’ <recnum)l(regnunr>(tnernst)
0043
tl'W
0045
Noteimerlisl and charlist are limited to six member end seven menbers
respectively.for reasons of practicality.
For the language to work,an objnase should have adefinition to be used.
Appendix B.
The Backus Naur Form Syntax of the KS's.
APPENDIX C. A COMPLETE LISTING OF THE CODE.
T h is
lis tin g
i s bound i n
a s e p a r a t e v o lu m e f o r s i m p l i c i t y , a n d rosy b e
o b t a i n e d fro m P r o f e s s o r M.G. R o d d , D e p a rtm e n t o f E l e c t r i c a l E n g in e e r in g ,
U n i v e r s i t y o f t h e W itw a te r s r a n d ,
1 J a n Sm uts A v e ., J o h a n n e s b u r g ,
Appendix C. A Complete Listing of the Code.
S o u th
Author Forsyth David Alexander
Name of thesis A paradigm for general scene analysis. 1985
PUBLISHER:
University of the Witwatersrand, Johannesburg
©2013
LEGAL NOTICES:
Copyright Notice: All materials on the U n i v e r s i t y o f t h e W i t w a t e r s r a n d , J o h a n n e s b u r g L i b r a r y website
are protected by South African copyright law and may not be distributed, transmitted, displayed, or otherwise
published in any format, without the prior written permission of the copyright owner.
Disclaimer and Terms of Use: Provided that you maintain all copyright and other notices contained therein, you
may download material (one machine readable copy and one print copy per page) for your personal and/or
educational non-commercial use only.
The University of the Witwatersrand, Johannesburg, is not responsible for any errors or omissions and excludes any
and all liability for any errors in or omissions from the information on the Library website.