Rochester Institute of Technology
RIT Scholar Works
Theses
8-1-1994
Comparing the ability of subjective quality factor and information
theory to predict image quality
Shyi-Shyang Li
Follow this and additional works at: https://scholarworks.rit.edu/theses
Recommended Citation
Li, Shyi-Shyang, "Comparing the ability of subjective quality factor and information theory to predict image
quality" (1994). Thesis. Rochester Institute of Technology. Accessed from
This Thesis is brought to you for free and open access by RIT Scholar Works. It has been accepted for inclusion in
Theses by an authorized administrator of RIT Scholar Works. For more information, please contact
ritscholarworks@rit.edu.
Comparing the ability of Subj ective Quality Factor and Information
Theory to predict Image quality.
By
Shyi - Shyang Li
B.S. Chinese Culture University
( 1982)
A thesis submitted in partial fulfillment
of the requirements for the degree of
Master of Science in the
Center for Imaging Science in the
College of Imaging Arts and Sciences of the
Rochester Institute of Technology
August, 1994
Signature of Author: _S~h_y_i-S_h_y_a_n_g_L_i
Accepted by:
Dana G. Marsh
_
>
~
Coordinator, M.S. Degree P gram
/
~ Ifff
.
COLLEGE OF IMAGING ARTS AND SCIENCES
ROCHESTER INSTITUTE OF TECHNOLOGY
ROCHESTER, NEW YORK
CERTIFICATE OF APPROVAL
M.S. DEGREE TIIESIS
The M.S. Degree Thesis of Shyi-Shyang Li
has been examined and approved
by the thesis committee as satisfactory
for the thesis requirement for the
Master of Science Degree
Dr. E. M. Granger Thesis Advisor
Dr. Dana G. Marsh
Mr. Joseph Altman
r
~/97'jI
Date
THESIS RELEASE PERMISSION FORM
ROCHESTER INSTITUTE OF TECHNOLOGY
COLLEGE OF IMAGING ARTS and SCIENCES
Comparing the ability of Subjective Quality Factor and Information
Theory to predict Image quality.
I, Shyi - Shyang (Robert) Li., hereby grant permission to the Wallace
Memorial Library of the Rochester Institute of Technology to reproduce
my thesis in whole to in part. Any reproduction will not be for commercial
use or profit.
Signature:
Date:
_=--
_----l....
_
_
ill
"
Comparing the ability of Subjective Quality Factor and
to predict Image
Information
Quality."
By
Shyi-Shyang Li
Submitted to the Center for
in
partial
fulfillment
for the Master
of
Imaging Science
of the requirements
Science Degree
Rochester Institute
of
at the
Technology
Theory
ABSTRACT
The
Information
methods of
the
is to
purpose of this project
Theory
This study
to predict image quality
exposing film. One
other allows
compare the
the flux to
exposure
ability
as a
of
the
function
holds the total
Subjective
of
film
Quality
speed
Factor
and
for two different
number of photons constant and
vary.
will:
1 Determine the relationship between
.
when
the resulting images are reproduced at the
2. Compare the resulting granularity
with
film
grain size and
the normal exposure method
and
image quality for
of the constant
flux
condition
the shutter speed as a function of
speed.
3. Compare the ability
accurately
predict
of
Subject
Quality Factor and
resulting image
quality.
flux
same size.
image quality
of varying
constant
Information
Theory to
ACKNOWLEDGMENTS
This
paper would not
have been
possible without the support of a
number
of people.
First,
to
insight
periods.
Dr. Edward M. Granger
and
guidance
Dr. Granger
Special thanks
served as
A
are
Thanks for
special
word
Thoughout the
on
principal advisor.
my thesis
I
must also thank
committee member as well as
Dana G. Marsh for the continuing
through the highs and lows that went with this
extended
being
of
long
the
ongoing technical
through the difficult
support.
support and encouragement
project.
the
provided
that allowed me to proceed
Mr. Joseph Altman for serving
providing technical
who
to Dr.
understanding.
thanks
is
owed
to
my
loving
period of this project she stood
wife,
steadfastly
Liang-Jen.
by
me.
The
many hours and days she willingly gave for me to complete this project.
Her unwavering patience made this struggle an enjoyable journey and I
will
forever be indebted.
Finally, I like
to
present
this
work
to
my parent, they
provided
and emotional support made the completion of this work possible.
VI
financial
TABLE OF CONTENTS
CERTIFICATE OF APPROVAL
ii
COPYRIGHT RELEASE
iii
ABSTRACT
v
TABLE OF CONTENTS
vii
LIST OF FIGURES
ix
LIST OF TABLES
x
I. Introduction
1 1
-
.
1
Granularity and Image Quality
l-2.Subjective
1
Quality Factor (SQF)
1-3. Information
4
Theory (I. T.)
7
1-3-1 General Information
1-3-2 Photographic Applications
1-3-3 Photographic Applications
7
:
Discrete Signals
8
Continuous Signals
11
15
n. Methods
2-1 Photograph
2-2
15
preparation
determining exposure condition
2-1-1
Setting up
2-1-2
Developing film
Evaluating image
and
and photographic paper
quality
2-2-1 Instructions to
33
2-2-2 Results from evaluating image quality
Calculating MTF
2-4
Determining granularity
29
32
of photographs
observers
2-3
15
of photograph
34
36
of photographs
38
Vll
TABLE OF CONTENTS (CONT.)
2-5
Calculating
of photographs
39
2-6
Calculating Information Capacity of photographs
40
SQF
III. Results
3-1 MTF
3-2
41
Granularity analysis
3-3 SQF
V
46
47
of photographs
3-4 Information
IV
41
of photographs
Capacity of Photographs
49
Discussion
51
4-1
Comparing the
4-2
Comparing the results
results
base
on
on
SQF
and
T-MAX-400
References
Information
and
Theory
51
TRI-X 400 films
54
56
VI. Appendices
58
Vlll
LIST OF FIGURES
FIGURE 1
Subjective
FIGURE 2
A typical
FIGURE 3
MTF
of T-MAX
1 00 film
w/
25
mm
lens
38
FIGURE 4
MTF
of T-MAX
1 00 film
w/
25
mm
lens
42
FIGURE 5
MTF
of T-MAX
400 film
w/
25
mm
lens
42
FIGURE 6
MTF
of T-MAX
3200 film
FIGURE 7
MTF
of TRI-X
400 film
w/
25
mm
lens
43
FIGURE 8
MTF
of TRI-X
400 film
w/
55
mm
lens
44
FIGURE 9
MTF
of T-MAX
400 film
FIGURE 10
MTF
of T-MAX
3200 film
FIGURE 11
SQF
FIGURE 12
Information
Theory and
FIGURE 13
Comparison
of
2
Quality Loss Vs Granularity
visual system
5
MTF
w/
w/
25
50
w/
mm
mm
135
lens
43
lens
mm
44
lens
45
48
results
SQF ranking
Granularity by using I.
IX
prediction
T.
and
SQF
51
53
LIST OF TABLES
TABLE 1
List
TABLE 2
Information
TABLE 3
Conditions
TABLE 4
Statistic data for
"Grey
TABLE 5
Statistic data for
"Egg"
35
TABLE 6
Statistic data for
"Flower"
36
TABLE 7
Granularity results
46
TABLE 8
Calculated
47
TABLE 9
SQF
TABLE 10
Results
TABLE 1 1
Information
TABLE 12
Results
TABLE 13
Summary
of
data
10
capacities of four
used
for this
films
project
Card"
14
30
35
Granularity results
47
results
of prediction
by using
SQF
Theory results
of prediction
by using Information Theory
of the results
48
49
50
52
I.
1
.
INTRODUCTION
Granularity and Image Quality
When
an emulsion
is uniformly illuminated
fluctuations due to the
This is known
unpredictable
negative.
as
random
distribution
photographic
of great
defined
and
determines the
density
and
statistical characteristic of
distribution
distribution
scanning,
of
with a
the
aperture
It is
that
in
blackening
each
a
of
in the microdensitometer, the
emulsion, and the type
image quality
will
be degraded
in image quality
study \Lisson,\ 983], the loss
variance of the noise
of emulsion
depends
and
(rms granularity),
It is
expressed
measured
in
the
of
sets a
)
is
limit
are
a well
negative,
is
on
[fT/.se/-,1986]
found to be linear
by
measured
a
have been
function
of
of
the
.
increases. In
density units and
the size
distribution
density
as the grain noise
was
small,
a measure of the random
image,
The granularity
a
exposed
the photosensitive material that
density.
emulsion.
area
fluctuation
uniformly
is
density
density fluctuations
density
Granularity
in the
small
science, because it
imaging
distribution in
photographic grain
a uniform
used
photographic
obvious
grain
in the
is to introduce
effect
root mean square
microdensitometer, areas
developed to
developed
by
silver particles
the grains in the developed photosensitive material.
exposed and
circular
the
(
the negative shows
Although the individual
photographic noise-level.
created
The
importance in
mean magnitude
developed,
developed
photographic
to the quality of photographic images.
unpredictable, their
of
granularity.
uncertainty into the
This uncertainty is
and then
a recent
with respect
shown
to the
in Figure(l).
Figure 1 Subjective
Quality Loss vs. Granularity
^r
o
z
->
m
O
_i
2-
3
a
m
>
3
tn
o
RMS
In this project,
optical system.
constant
for
means
samples will
be
be
pupil
from
printed
a given area on
the original object.
speed, the lens focal length
Since the final
and
prints made
the
to
obtain equal print size.
will pass
As
image
through
a
a result of this
size
increase in
from different films
to the same size, the system MTF and
magnification required
flux in the
exposed under constant photon
that in a given amount of time, N photons
to the film speed.
speeds will
by the
film
a constant shutter
proportion
different
scaled
This
diameter lens
restriction
direct
some of the
Granularity
granularity
with
will
be
The
following
diagram
provides
information
about
the relationship
of
the
object
lens to
the different size images that result due to film speed.
25
mm
lens
F#:4
Slow speed film
50
mm
lens
F*:8
135
mm
lens
F#:22
Fast
speed
film
As
shown
in the
the
image
generated.
diagram,
the same size print
as
with
The
the constant flux condition, the faster the
negative must
Quality Factor (
A
difference in image quality
SQF
phenomenon of just noticeable
change
in the
judging
weight and
Granger
to make the
slower
film
generate
)
loudness
[Granger, 1972]
can
difference
has been
stimulus
to a logarithmic spatial
be
(JND) being
for many
observed
follow
of sounds
by changing
obtained
what
is known
by 7-10%.
This
to a constant percentage
related
neural
SQF
processes.
as
Tasks
Weber's Law.
such
as
This led
to hypothesize that image quality might in some way be related
frequency
(OTF). If so, image quality
spatial
magnified
the bigger
the fast film.
2. Subjective
perceptible
be
film,
weighting
of
the
system
Optical Transfer Function
might correlate with the area under the system
OTF
on a
log
frequency scale.
The Subjective
Quality
result of a search
directly measured
for
in
Factor
(
SQF
an objective
)
was
figure
developed
by
Granger
of merit which could
and
be easily
Cupery,
as
the
calculated and
practice and which would correlate with subjective rank regardless of
Modulation Transfer Function
(MTF) form.
A
number of experiments were performed
to
test the quality factor for a wide variety of MTF shapes. The results of the experimental
program were
was
0.988
that SQF
was able
correlated with
the
to predict
measured
data
image quality
within normal reader error and
[Z,/55o,1983]
.
The quality
of a visual
visual system
typical
has
an
visual system
image is
MTF
related to the scale of the
which peaks
MTF is
in the
shown plotted
Figure 2. Atypical
Spatial
postulating
stimulate
Based
on
system
this
effect
which
MTF( including
scaled
includes the
the retina. A
2.
MTF
100.0
Cycles/Degree
nature
for the eye, the limits
of
integration to
have been defined.
these observations, a
MTF has been
-
the retina. The human
cycles/mm at
10.0
Frequency
arbitrary bandpass
an
10-20
on
log frequency in Figure
visual system
1.0
0.1
By
Vs
region of
image
eye.
lenses
to the
one-dimensional
and
films)
retina of
SQF
was
between the limits
the observer
by
defined
of
as the
10-40
integral
of
the
cycles/mm when
the
the magnification of the system,
40
SQF
=
k\\T(f)\d(\ogf)
(1)
10
r(f) is the Optical Transfer Function.
/ is the
\/k is
spatial
a
normalizing
integration to
There is
no
Because
real
by equating the above
=1.
images involve
system.
weighting information
by describing
constant obtained
to limit the considerations to
reason
factored into the
MTF
frequency.
a
That
over all
the
two-dimensional
system
is,
MTF,
the impression
directions. The SQF
MTF in
one-dimensional
a
two dimensional MTF must be
quality is
of
MTF descriptions.
value can
polar coordinates and
obtained
by
for
a general
performing the
following
be
obtained
integration:
402*
SQF
=
kj\\r(f, ep{loS f)W
(2)
10 0
Where:
/ is the
spatial
frequency in
cycles
/
mm
along
of line structure.
it is the
appropriate
normalizing
constant.
402*
A
=
JJ<5(log/)<a?
10 0
(3)
equally
a given azimuth
9
The
above
the limits
It is
formula
of the
is
not
magnification.
when
Image Sharpness Scale
characteristic
assessment
allows a simple calculation of the
the
of
(ISS) [Granger, 1972]
system
and
of
subjective
intrinsic to the image itself but only
Therefore, it is important
calculating the SQF
3 Information
SQF
quality
of a print which
within
a
quality
range.
quality
evaluation, that
image
to the
lies
as viewed at a specified
that the proper system magnification be specified
of a system.
Theory
3-1 General Information
The theorems
the
within
general
Shannon in 1948. These theorems
communication
communication
channels
system,
field
were
of
information theory
developed
largely
but they may be readily
such
are
within
based
by
the context of electrical
to
adapted
on research
any
other
as the optical transmission or photographic
type
of
recording
of
information.
It is
estimated
per year and
that the total
that this
amount of printed
figure doubles
information
about each
decade. In
future large-scale information-handling problems, the
approach as provided
by information theory is
alone
need
self-evident.
is in
1016
excess of
view of
for
a
bits
these present and
fundamental
analytical
In
is
photographic applications the photographic process
it is important to
and
achieve the
incoming
structure of
the
how to best
present the signal to
be desirable to
in the
photographic
system
noise.
is
When
a
specified
photographically
the
known,
well
are
This
determined
rate.
question
element.
type
the
of
and
signal
(for example,
highest information
M
=
N
close
result, it may
frequencies
spatial
and
spectrum of
relationship between
binary form)
storage per unit
may
log2
KG2
M
+
Altman
in the image using
[Z,ev7,1958]
.
(4)
1
is to be
(5)
stored
area, simplified models
prove adequate.
:
-^r
2
in
on a unit storage cell
in the form
=
a
Discrete Signal
according to Levi
C
As
one of
[Da int y&Shaw,l 974].
DQE
for the influence
written
statistical
becomes
the system MTF and the Wiener
based
of noise
The
medium
highest information capacity
demonstrates the very
approach
application :
by
gave a method of analysis
be
the
the photographic recording
photographic process as a storage medium
unit area can
and
element which yield the
recording
with
highest information recording
fairly
information capacity, recording rate,
3-2 Photographic
the recording
the Wiener spectrum of incoming signals to the
These frequencies
rate.
the
match
signal
used as
and
of
Zweig
a simple model
The information capacity
per
where: c
is
a constant.
C is the information capacity
M is the
The
for
parameter
number of cells
R is the
density range (D^
A is the
cell size.
R may be
a specified separation
written as equation
A~'
be
assumed to
criterion,
constant
only A
K,
for
levels
are
of equation
for
capacity.
a series of Kodak
(5)
can
be
be
as
(6):
(6)
films
A
I
cA2
(6)
+1
that C increases as A
information
we conclude
This
[^]log2
=
reveals
maximum
necessary, so
information
a given photographic process and so
remains as a variable and equation
(
small as possible
D^J
-
=
C
Investigation
levels
number of recording
N is the
N
of the channel
conclusion
is
confirmed
which are summarized
However,
capacity.
that, in principle,
decreases,
binary
by the
in Table I.
at
so
A
should
least two recording
recording
will give optimum
results of Altman and
Zweig
for
TABLE I
Spread
Available
function
levels
Bit capacity for an image
of
0 x 10
area
diameter
fjm
logf
M
Kodak Fine Grain
M-level
Binary Experimei
8
3
0.33
0.11
12.5
6
2.6
1.6
0.64
1.1
15
4
2
0.88
0.44
1.1
3
1.6
0.33
0.21
15
3
1.6
0.7
0.44
0.5
27
2
1
0.14
0.14
0.05
1
2
1
160
160
160
Altaian
and
Cine Positive
Recordak Fine
Grain Type 5454
Recordak Fine
Grain Type 7456
Kodak
plus-X
Kodak Pan-X
lodak Royal-X Pan
Kodak High
Resolution Type
649
An
analysis
of
binary
and
[Altman& Zweig,\ 963] concludes
rather
in M
than
gives
binary recording is not
only
a
multilevel
that the gain in
substantial.
cell size or spread
function
10
information capacity
It follows from
logarithmic increase in capacity,
important factor. The
by
recording
and
equation
therefore
having a
by using
Zweig
multilevel
(4) that an increase
small cell size
storage area
is the
most
of\fjmx\/jm,
as
opposed to
100/imx
100/xw,
messages and codes are
coding
be
For
advantage over
increase in capacity
104
Where
of
binary
commonplace, recording levels higher than two may involve
difficulties,
complications and
well separated.
would give an
especially in
view of
the fact that the levels have to
these reasons multilevel recording may
all
offer
little
practical
binary.
3-3 Photographic Application: Continuous Signal
It is important to know that the
continuous approach will not
they may
allows
or
the information capacity to
apparent.
For
photographic
an
same as
types of information
turn the close relationship
for
be the
using
applications
of the
signal, this
be
input. The
expressed as a
in
scientific
spatial
must
photography
frequency
approach
continuous channel with average mean-square
photographic
process
fog density
is
more
and
photographic process and
great
nearly
Dmax. A
function
the wide
a peak-limited
variation of
operating limits.
11
and
practice
concern
frequency,
systems,
is that
and
including
highest information
as the result
Gaussian
channel,
with
noise.
in
its operating
over
the
rate
for the
In fact the
due to the non-linearity
imaging properties
a
DQE then become
and
is usually taken
arises
its
rate
they
using
of this approach
of spatial
achieve the
limitation
difficulty
benefit
where overall
be designed to
when
discrete approach, but in
between information transfer
recording element,
incoming
between
for information capacity
turn out to be quite close. The results are different because
different questions,
it
results obtained
region
of
the range
the
of its
According
to
Dainty [Da int y& Shaw, 197 4]
continuous channel with an average
C
If the
B,
=
P
Af log2
+
C
If we
use
noise
WN(f)
the
1/2
=
limitation is
:
N
(7)
N
signal and noise power are
each of width
"power"
the result for the information capacity of a
f,
then the
approximately
constant over two adjacent regions
capacity for the total band
|A/(log2(l+
power spectrum of
P/N)
+log,(l +
the signal,
width approximates
%J)
WN (/) for
P
and
A
and
to :
(8)
the power spectrum of the
for N, then
f
flog2
Since
de
signal and noise are
two dimensional functions of space, for photographic
77
Since the
assumed
(
statistical properties of
to be
isotropic,
(9)
wN{f\
V
Ws(u,v)
the
and since
dudv
photographic
for
optimum
12
process,
coding the
images
(10)
including
image noise, may be
signal will also
have the
nature
isotropic
of an
spatial
pattern, it is
noise
frequency,
w2
w,
where
convenient to work
u2
v2
+
=
,
in terms
(H)
wdw
If we
assume that a natural scene
C
r
x
=
since
Equation
(11)
Ws(w)
constant over a
as
an attempt
to
constrained
theorem
for
2/,..\A
( 1+ MTF2(w)
\>
d>v
limited
calculate
between
the
(12)
w
evaluating the information capacity
of
input/output,
restrict
it to
maximum
fog density
ratio
in terms
exposure/density
Dmax, Jones
of
the
of power spectra will
range.
To
keep
equation
is
peak-limited.
13
the photographic process
[./owes, 1961]
then made various ad
photographic process
of
small signals condition.
information capacity
and
a power-limited channel and
for the fact that the
to 1/w then
MTF2(w)
Due to non-linearity the S/N
(1 1) "exact", it is necessary to
In
a power spectrum proportional
illustrates the dilemma
photographic process.
only be
\og2
has
the one-dimensional
leading to
WN{w\
V
o
of
By
hoc
usi
used
Shannon's
corrections to account
for the
of
information
respective
interest,
are summarized
Table II. Information
estimated
Film
capacities.
His results, along
with other comparative values
in Table II.
capacities of four
films,
and various comparative
values,
as
by Jones.
Information
Area for 1
Information
capacity
bit
rate
Comparative
Exposure
for
one
bit
Film
time for Hi-Fi
area
equiv.
system
one
to
TV
frame
bits
2
fjm2
cm
bits erg
(xia-)
'
photons
K")
(x,0-)
sec cm
2
em2/
/frame
R.oyal-X
0.449
200
26.5
8.18
3.01
2.98
Tri-X
0.845
118.4
7.35
29.4
5.1
1.76
Plus-X
1.86
53.8
6.45
33.6
11.2
0.8
Pan-X
2.85
35.0
7.45
29.2
17.2
0.52
Although
manufacturers of
image quality is
given.
film
provide a speed
While the
speed
insufficient
ordinary photographer, it is
the
We have defined Information
capacity
of a
film to
Theory
receive and store
each
rating is usually
when
greatest possible amount of information
rating for
choosing
has to be
a
14
usually
satisfactory
film for
recorded
and we will use
information.
a
film,
no
rating
guide
of
for the
scientific purposes where
by the film.
this powerful tool to obtain the
n
The
experimental
2. 1 Photograph
2.1.1
setup
and
Four films
were used
T-MAX 3200
all
and
four films. It
following steps:
in the study, they
Kodak T-MAX
"older"
an
are:
medium
be
would
grain,
and
interesting to
T-MAX 3200 has the largest
know how they
we
professional
films
product of Eastman
actually take
combination and also
illustrate the
variables
IM
grain of
perform under normal
flux.
are newer
Kodak
a picture of the
the
shutter speed.
needing to be
15
so
and
it is
TRI-X 400 is
interesting to find if
products.
object,
The
we must select
the lens and film
following equations
controlled
(13)
LS
products,
Company,
is any difference between these two
Before
100, T-MAX 400,
T-MAX
TRI-X 400. T-MAX 100 has the finest grain, T-MAX 400
exposure and constant
(F-y
the
determining exposure time.
TRI-X 400 have
there
on
preparation:
Setting up
and
is based
process
METHODS
in the
experiment.
are used
to
t : shutter speed
L
luminance in
S
film
:
cm2
cdls/
speed
F#
: numerical aperture
F#=f
where
I D
(14)
f : lens focal length
D
When the
:
lens diameter
shutter speed and
relationship
can
be
the total flux are constant, the
following
established:
4ASA
If we let
and
a
order
same,
>
when
ASA=100,
ASA=3200.
a proper range of focal
25 mm,
f,
=4 when
F#=22;
have
In
F*
focal length
Also,
>
According to
D
of 48 mm when
to have constant
it follows that F#=8;
must
be
constant
length. When F#=4
flux,
the
while new shutter speeds
h
then
t3
F#=8,
and
t2
we use a
and a
shutter speed
are set
(i.
e.
lens
(f,)
ASA=400
D=6 mm) in
focal length
setting
for
when
of focal
of
132
needs to
order
length
whenF#=22.
be the
normal exposure and
tr
the
above
discussion,
the
following
determined:
16
to
exposure conditions can
be
(1) For T-MAX100 film:
(a)
A lens
of 25 mm
focal length
F#
and
of 4 was used
to take a picture of each
object.
(2) For T-MAX 400
(a)
A lens
of
50
film:
F#
focal length
mm
of
and
8
was used
to take a picture of each
object.
(b)
A lens
of 25 mm
object.
This step
F#
focal length
and
produced the same
of 4 was used
image size, in
to take picture
order
of each
to study the effect
of different magnification on each print.
(3) For TRI-X 400 film:
F#
mm
focal length
and
(b) A lens
of 25 mm
focal length
and
object.
This step
(a)
A lens
of
50
of
8
was used
to take a picture of each
object.
produced
F*
the same
of 4 was used
image size, in
to take
order
a picture of each
to study the effect
of different magnification on each print.
(4) For T-MAX 3200 film:
(a) A lens
of
135
mm
focal length
F#
and
of 22 was used
to take a picture of
each object.
(b)
A lens
object.
of 25 mm
focal length
This step
produced
F*
and
of 4 was used
the same image size, in
to take a picture of each
order
to study the effect
of different magnification on each print.
To
summarize
the
above statement
the
following combinations
our project:
17
have been
used
for
(1). Conditions to
produce constant
flux (fixed
F#
Focal Length
amount of photons):
Shutter
speed
T-MAX 100
25
mm
4
1/30
sec
T-MAX 400
50
mm
8
1/30
sec
TRI-X 400
50
mm
8
1/30
sec
135
mm
22
1/30
sec
T-MAX 3200
(2). Conditions to
give normal exposure:
F#
Focal Length
T-MAX 400
25
TRI-X 400
25
T-MAX 3200
(3). The
25
mm
mm
"busy"
Photographs
Shutter
speed
4
1/250
4
1/250
sec
1/1000
se
4
mm
orders of the pictures
gray card,
(r,)
(/,)
sec
taken for the scenes are : gray card,
scene.
attached:
18
"simple"
scene,
V
Ima<?e from T-MAX 100 film / 25
Image from T-MAX 400 film / 25
19
mm
mm
lens
lens
20
Image from TRI-X 400 film / 25
mm
lens
Image from T-MAX 3200 film / 25
mm
lens
Image from T-MAX 400 film / 50
Image from TRI-X 400 film / 50
21
mm
mm
lens
lens
22
Image from T-MAX 3200 film / 135
mm
lens
Image from T-MAX 100 film / 25
mm
lens
Image from T-MAX 400 film / 25
Image from TRI-X 400 film / 25
23
mm
mm
lens
lens
Image from T-MAX 3200 film / 25
Image from T-MAX 400 film / 50
24
mm
mm
lens
lens
25
Image from TRI-X 400 film / 50
mm
lens
Image from T-MAX 3200 film / 135
mm
lens
Image from T-MAX 100 film / 25
Image from T-MAX 400 film / 25
26
mm
mm
lens
lens
27
Image from TRI-X 400 film / 25
mm
lens
Image from T-MAX 3200 film / 25
mm
lens
Image from T-MAX 400 film / 50
Image from TRI-X 400 film / 50
28
mm
mm
lens
lens
Image from T-MAX 3200 film / 135
2.1.2
Developing film and
(1)
All films
were
photographic paper.
developed
Film Processor V-5N for
(2)
at the
RI.T. campus, using the Kodak Versamat
developing the films.
(a)
Process
speed
for T-MAX 100
was
2.2 ft /
(b)
Process
speed
for T-MAX 400
was
2.75 ft /
(c)
Process
speed
for TRI-X 400
(d) Process
speed
for T-MAX 3200
The
was
1.5 ft/
was
min.
min.
min.
2.2 ft /
min.
following conditions were used to project images from negative to
photographic paper:
29
mm
lens
Table ILL Conditions
(1) When 25
mm
#,& H2:
Films
lens
371/2"
used
for this
project
was used:
&
Objects
53/4"
Filter
Exposure Time
X
Magnification
T-MAX 100
Gray
T-MAX 100
Card
3.5
27
sec
631.9
Egg
4.0
44
sec
760.8
T-MAX 100
Flower
4.0
48
sec
860.4
T-MAX 400
Gray Card
4.0
38
sec
608.8
T-MAX 400
Egg
4.5
27
sec
597.3
T-MAX 400
Flower
4.5
27
sec
632.2
TRI-X 400
Gray Card
3.5
28
sec
601.4
TRI-X 400
Egg
4.0
46
sec
784.0
TRI-X 400
Flower
4.0
46
sec
632.2
T-MAX 3200
Gray
T-MAX 3200
Egg
4.0
31
sec
784.0
T-MAX 3200
Flower
4.0
33
sec
632.2
Card
37
4.0
30
sec
711.5
(2) When 50 mm lens was used:
//,& H2:
Films
371/2"
&
Objects
53/4"
Filter
Exposure Time
X
Magnification
T-MAX 400
Gray Card
T-MAX 400
Egg
T-MAX 400
4.0
11.5
sec
190.4
4.0
10.0
sec
196.0
Flower
4.0
10.0
sec
215.1
TRI-X 400
Gray Card
3.5
9.0
TRI-X 400
Egg
3.5
7.5
TRI-X 400
Flower
3.5
(3) When
135
mm
H,& H2:
lens
196.0
sec
7.0
215.1
sec
was used:
371/2"
53/4"
&
Exposure Time
Filter
Objects
Films
190.4
sec
x
Magnification
T-MAX 3200
Gray Card
T-MAX 3200
Egg
T-MAX 3200
Flower
(3). All
photographic paper was
7.7
sec
21.1
4.0
6.3
sec
21.5
4.5
7.1
sec
21.4
4.5
developed
at
the R.I.T.
Kreonite B/W Process
(4). All
negatives were printed
to the
same
image
31
size.
campus
by using a
2-2
Evaluating image quality of photographs
The
successive categories method was used
The underlying
stated
(1)
The
assumptions of the
by Togerson (1958)
law
in this
project
for
statistical analysis.
of successive categorical
[Togerson,\95S]
judgments have been
:
psychological continuum of the subject can
be divided into
a specified number
of order categories or steps.
(2) Owing to various
necessarily
and
always
sundry
located
also projects a normal
factors,
a given
category
boundary is not
at a particular point on the continuum.
distribution
of positions on
different category boundaries may have different
the continuum.
mean
locations
Rather, it
Again,
and
different
dispersions.
(3)
The
subject
whenever
judges
a given stimulus to
be below
a given
category
boundary
the value of the stimulus on the continuum is less than that of the
category boundary.
There
are
many forms
techniques and
data
of category
scaling
reduction algorithms
and a wide
about the
participated
image quality
in this
photograph on a
project.
7-point
scaling
was used
of twenty-one photographs.
They were asked to
scale.
The instructions
32
of experimental
that have been used in category scaling. A
common experimental method of category
data
variety
in this
project to gather
Thirty-three
rate the overall
observers
image quality
and results were as
follows:
of each
2-2-1 Instrustions to
observers
INSTRUCTIONS TO OBSERVERS
You
will
be
the image quality
Please do
Do
of the
not
photograph,
dirt,
and
1
intervals
any
physical
rating for the
defects in the
should not exceed
express your opinion
unusable and
using
you
to
make a
judgment
on
print.
represents excellent
of image quality.
The
image
quality.
categories used
(2) Very Good
Good
(4) Acceptable
(5) Unsatisfactory
(6) Poor
(7) Unusable
33
photograph.
12 inches.
a scale of number
(1) Excellent
(3)
like
not consider composition.
The viewing distance
equal
and give a
would
directly touch the photographs.
Ignore scratches,
Please
We
shown a number of photographs.
from 1 to 7
where
Numbers between 1
in these
7
and
experiment are:
7
represents
represent
You may
not use
from 1 to
7;
fractions
no other
or
decimals;
integers may be
you must use
An image quality
three months.
professional
were
assessment
The
presented
of the photograph
is
randomly chosen, among them
and
ordinary
observers.
The
were:
photographs
to each observer for evaluation. The randomness
is important. This
performed
process allowed control of accuracy of
to generate
mean value
(2) Thirty three viewers were asked to make
(m)
viewers a statistic
and standard
a judgment on
deviation
photograph and a
means
(3) Data:
The
gray
card:
rating
and
excellent,
"7"
was given
(S)
the image quality of
"
the
be
of photographs
the rating data. After collecting all the data from the
analysis
should
by the observers was performed in a period of
observer were
people, students,
randomly
The integers
used.
2-2-2 Results from evaluating the image quality
(1)
integers.
to the photograph. A rating of 1
"
means unusable.
Appendix 1
abbreviations are as
Egg:
G,
T-MAX100 / 25
TRI-X 400 / 25
follow:
mm:
mm:
T-MAX 3200 / 135
T-MAX 400 / 25
1,
TRI-X 400 / 50
4,
mm:
Flower: F
E,
7
34
mm:
mm:
5,
2,
T-MAX 3200 / 25
T-MAX 400 / 50
mm:
mm:
3,
6,
(4)
A
(5)
The
statistical analysis was performed
original results were
represented
identified
as
by
1
to evaluate the data.
transferred to a
and original
7 is
new scale where
represented
by 0.
These
the
original
1 is
new numbers are
"Ranking"
Ranking (R)
=
116.666-16.666
x
mean
(m)
(15)
Table TV Statistic data for
"Grey
Card"
Gl
G2
G3
G4
G5
G6
G7
3.55
4.15
5.30
4.52
3.0
2.30
1.79
Ranking (R)
0.575
0.475
0.283
0.413
0.666
0.783
0.868
Standard Deviation
0.99
1.08
1.34
1.13
0.85
0.90
0.73
Mean
(m)
w
Table V Statistic data for
Mean
(m)
Standard Deviation
(S)
"
Egg
"
El
E2
E3
E4
E5
E6
E7
3.97
4.73
5.36
4.97
3.36
2.42
1.88
1.03
1.11
1.20
1.11
1.07
0.99
1.01
35
Table VI Statistic data for
Mean(m)
Standard Deviation
2-3
(S)
Calculating MTF
Many
F2
F3
F4
F5
F6
3.52
4.52
5.39
4.97
3.18
3.21
1.91
0.93
1.13
1.13
1.19
1.03
1.01
0.71
The quality
The human
response at
of
eye
MTF is
cycles per
used
one-dimensional
in the
includes the
of
logarithmic
log
spatial
obtained
image is
by
changing the
quality
In
rank.
proven quite successful
function
rank can
fact,
that a perceptible
scale of
related to the scale of the
a modulation transfer
These
noticed
(MTF)
be
Quality
frequency
frequency
weighting
(SQF)
of
correlates with
the
tells
computed
computations
scale.
36
that
the retina.
broad
if the
use
rank
peak
"true"
only the
for
two-
that the one-dimensional treatment
us
under
visual properties
that image quality
system optical
the area
on
with
in predicting quality
successes suggest
Factor
the point spread
image
two-dimensional
weighting function to describe the
Specifically, image quality
on a
be
calculation of
Subjective
spatial
has
structure.
proper
the image.
can
degree (cpd). Image quality
MTF have
dimensional image
image quality definition have
of a visual
visual system
6
F7
of photographs
difference in image quality
function.
"
Flower
Fl
in the field
studies
"
the
transfer
system
is
related to
function (OTF).
OTF
when
displayed
A Crosfield Magnascan 636
All
card photographs.
The data
MTFs'
4*)
=
drum
photographs were
were then read
calculation.
reflection
in
Photoshop
scanner was used
carefully
4-5 to
scan seven of
aligned and scanned at
generate raw
of each print were calculated
to
pixels
data for granularity
according to the
^
18
following
the gray
/
and
mm.
MTF
equations:
d6)
(17)
md07F(f)=]i(x)ea'*'&
oo
MTF{f)
A
routine was written
calculations are shown
The MTF
curve
combination
is
as
=
\OTF(f\
in Mathcad 4.0
(18)
and used to
do the
calculation.
The
program and
in Appendix-3.
for the
print
generated
from T-MAX 100 film
following:
37
and
25
mm
lens
Figure 3 MTF
of final print
by using T-Max
1.5
100 film
w/25 mm
lens
2.5
2
Frequency (lines/mm)
2-4
Determining granularity
Photoshop 4-5
obtained
with a
The digital data
used
in the
low
obtain
filter
pass
and
were scaled as an
equation
granularity data to
^New
to
was used
below. The
"New"
log
one set of data was
the granularity
data,
the
data was
second set
integers between 0
and
255. A
mean of
128
filter.
was
following equation was used to transform original
granularity data:
Noise(p)
White
obtained without a
+
log2
,
38
(19)
No^y-hlfpl
V
where white
A linear relationship
pixels
=
was
255
developed between Information Theory
predict the subjects averaged response
was
relationship
response
2-5
for
Calculating
objective
for
developed between SQF
A
each set of prints.
and
oD to
predict
and
aD to
second
linear
the subjects averaged
each set of prints.
SQF
A Subjective
(20)
vofpixels
of photographs.
Quality Factor (SQF) was developed
figure
of merit which could
be easily
as the result of a search
calculated and
the eye
merit
function
image
appearance
averaged response
granularity.
was
for
developed between SQF
each set of prints.
form.
linearly when the actions of
including the magnification of the image are taken into
A linear relationship
and
aD to
Image quality is
consideration.
predict the subjects
related
to both MTF and
They act independently as when increase in granularity then we expect
loss in image quality
quality
predicts
an
directly measured in
practice and which would correlate with subjective rank regardless of MTF
The SQF
for
and also when a
loss
as well.
39
of MTF we can expect a
loss
of image
a
/
A
Q.
=
SQF
routine was written
2-6
aaD
(21)
in Mathcad 4.0
calculations are shown on
by using
-
and used
to calculated SQF. Programs and
Appendix-4. An image quality
assessment was performed
SQF.
Calculating Information Capacity.
Information capacity
of an emulation
depends
(MTF) and the granularity of the emulsion.
obtained
and
first,
<jd to
I.Q.=
A
the modulation transfer function
In this study MTF
of each system was
then a linear relationship was developed between Information
predict
the subjects averaged response for
Theory
each set of prints.
a0+b0(lC)
routine was written
Capacity. Programs
was also used
granularity
on
in Mathcad 4.0
to calculate Information
and calculations are shown on
to predict image quality
into
and used
Appendix-5. Information
Theory
by calculating information capacity and taking
consideration.
40
m. Results
3-
1 MTF
All
(1)
of the
final
prints was calculated:
of the seven
gray
card photographs were scanned and
Crosfield Magnascan 636
and a
scanning
(2). MTF's
4x)
=
of
18
reflection
pixels
/
drum
scanner.
An
digitized
edge
trace
by using
was
performed,
mm was used.
of each print were calculated
according to the
following equations:
*x)
dx.
otfw)=
J(xy2^ac
n=l
MTF{f)
A
=
\OTF{fl
routine was written
are as
in Mathcad 4.0
and used
follows:
41
to do the
calculation.
The MTF's
Figure 4 MTF
of final print
by using T-MAX
2
1.5
100 film
w/
25mm lens
2.5
Frequency (lines/mm)
Figure 5 MTF
of final print
by using T-MAX 400 film w/25
2
1.5
Frequency (lines/mm)
42
"
mm
lens
Figure 6 MTF
of final print
by using T-MAX 3200 film w/25
2
1.5
mm
lens
2.5
Frequency (lines/mm)
Figure 7 MTF
of final print
by using
TRI-X 400 film
2
1.5
Frequency (lines/mm)
43
2-5
w/25 mm
lens
Figure 8 MTF
of final print
by using TRI-X 400
film w/ 50
mm
lens
mm
lens
Frequency (lines/mm)
Figure 9 MTF
of final print
by using
T-MAX 400 film
2
1.5
Frequency (lines/mm)
44
2.5
w/
50
Figure 10 T-MAX 3200 film
w/
135
mm
lens
0.98
0.96
0.94
-
0.92
-
0.9
0.88
-
15
2
2.5
Frequency (lines/mm)
3-2
Granularity analysis
Photoshop
4-5
was used
to obtain the granularity data from the final prints.
this measurement contain a systems level
and scanner used
measurement.
second set of
data
integers between 0
and
and
an
for
the
readings
from
One
of
MTF, including MTF
set of data was obtained
was obtained without a
255
Photoshop 4-5
and mean of
128
with and without
45
of
film,
with a
filter. The digital data
was used
the low
for
pass
calculation.
filter
are:
Therefore
paper, lens
low
pass
filter
were scaled as
Granularity
Table VII
Granularity results
T-MAX 100
T-MAX 400
T-MAX 3200
TRI-X 400
TRI-X 400
T-MAX 400
T-MAX 3200
25
25
25
25
50
50 mm
135
mm
mm
mm
mm
mm
mm
Print
Granularity W/ filter
5 1
9.3
10.3
8.0
4.6
5.4
5.1
Print
Granularity w/o
9 4
9.7
19.1
15.3
8.7
9.6
14.6
By using
.
.
equation:
New
'log^M)+,g2
White
V
we
have
new
granularity data
as:
Table VHI Calculated
Granularity results
T-MAX 100
T-MAX 400
T-MAX 3200
TRI-X 400
TRI-X 400
T-MAX 400
T-MAX 3200
25
25
25
25
50 mm
50 mm
135
mm
mm
mm
mm
Print
Granularity W/ filter
0.0183
0.0322
0.0353
0.028
0.017
0.0196
Granularity w/o
0.0325
0.0334
0.0486
0.0507
0.0303
0.0331
print
46
mm
0.0187
0.0621
3-3 SQF
of photographs
Image quality
was
determined
by using the correlation of SQF &
aDF
subject
data for
each photograph.
(1). A
routine was written
in Mathcad 4.0 to
calculate
Table TX SQF
The
results
T-MAX 100
T-MAX 400
T-MAX 3200
TRI-X 400
TRI-X 400
T-MAX 400
T-MAX 3200
25
25
25
25
50 mm
50
135
mm
0.57
SQF
(2)
the SQF.
mm
0.51
mm
mm
0.54
0.48
following equation was used to take granularity into
2
{l\Pr^on){N)-aSQF{N)-ba0{N))
=i
a=l andb=-3.6
then
R(prediction)
=
SQF-3.6G
47
=0
0.77
mm
0.79
consideration:
mm
0.97
Table X Results
(prediction)
R,
'(measured)
of prediction
by using
SQF
T-MAX 100
T-MAX 400
T-MAX 3200
TRI-X 400
TRI-X 400
T-MAX 400
T-MAX 3200
25
25
25
25
50
50 mm
135
mm
mm
mm
mm
mm
0.51
0.40
0.36
0.44
0.71
0.72
0.91
0.57
0.51
0.48
0.54
0.77
0.79
0.97
Fig
11 SQF
results
1
0.9
0.8
0.7
0.6
O 0.5
0.4
-
0.3
0.2
--
0.1
--
0
4
Films
3-4 Information
Capacity of photographs
Image quality
mm
was obtain
by using Information Theory for each print:
48
A
(1)
in Mathcad 4.0 to
routine was written
granularity taken into
calculate
The
consideration.
results are as
Table XI Information
I.C.
(2)
with
the
follows:
Theory results
T-MAX 100
T-MAX 400
T-MAX 3200
TRI-X 400
TRI-X 400
T-MAX 400
T-MAX 3200
25
25
25
25
50
50 1
135
63.2
63.5
mm
30.9
The
information capacity
mm
30.0
following
mm
24.8
equation was used
mm
31.1
mm
47.1
r
to take granularity into consideration:
%^M=00884+00119*IC
Table XII Results
R{prediction)
of prediction
by using Information Theory
T-MAX 100
T-MAX 400
T-MAX 3200
TRI-X 400
TRI-X 400
T-MAX 400
T-MAX 3200
25
25
25
25
50 mm
50 mm
135
mm
mm
mm
mm
0.46
0.44
0.38
0.46
0.65
0.84
0.84
0.57
0.51
0.48
0.54
0.77
0.79
0.97
D
(measured)
mm
49
TV DISCUSSION
4-1 Comparison
Judging the
of results
results
can produce
very
predicting image
using SQF
from Figure
and
(12)
reasonable results.
Information Theory.
we can
But
say both SQF
overall
the SQF
and
Information
does
a
Theory
better job
of
quality.
Figure 12 I.T.
and
SQF ranking
prediction
.
.
/
I.T.
s
SQF
_
il.
/
0.3
0.5
0.4
Ranking
-
06
Measured
50
The
prediction of the
Is Image
Quality
image quality Vs
a
granularity
(crD) :
?
,
-J
measured
ASA
(a) When constant flux is used for exposure, the fast film has better image
It is because the
fast films
are
gain
usually
(b) Under normal
in MTF
more
grain
effects; also, because the
better.
sensitized
exposure
than offsets the
quality.
the slower film
Table XIII
has better image
Summary of the
quality.
results:
T-MAX 100
T-MAX 400
T-MAX 3200
TRI-X 400
TRI-X 400
T-MAX 400
T-MAX 3200
25
25
25
25
50
50 mm
135
mm
mm
mm
mm
mm
r
SQF
0.57
0.51
0.48
0.54
0.77
0.79
0.97
Information Cap.
10.16
9.98
10.55
9.8
15.21
21.73
27.72
Granularity w/ fil.
5.1
9.3
10.3
8.0
4.6
5.4
5.1
Granularity w/o
9.4
9.7
19.1
15.3
8.7
9.6
14.6
0.51
0.40
0.36
0.44
0.71
0.72
0.91
0.46
0.44
0.38
0.46
0.65
0.84
0.84
^(prediction/
SQF)
^(prediction/ 1.
T.)
Note:
(1)
Constant flux
condition:
G5: TRI-X 400 film / 50
mm
G6: T-MAX 400 film / 50
lens
mm
lens
51
(2) Normal
exposure condition:
G2: T-MAX 400 film / 25
G4: TRI-X 400 film / 25
mm
mm
Figure 13 Comparison
lens
lens
of
granularity by using I. T.
and
SQF
0.05
.&
0.04
C
O
0.03
4
Films
Photoshop
4-5
was used
to obtain the granularity data from the final prints.
level
measurement contain a systems
scanner used
for
the second set
measurement.
of
integers between 0
compare
reach
the
data
and
One
MTF, including MTF
set of
data
was obtained
was obtained without a
255
and a mean of
the granularity on each
following
of
by
using SQF
52
with a
for the
was used
conclusions :
film,
and
paper, lens and
low
filter. The digital data
128
photograph
of
Therefore this
pass
were
and
scaled as
When
we
Theory,
we
calculation.
Information
filter
(a) When
SQF is
Under
for
used
prediction:
flux the granularity in the
constant
normal exposure
aD
varies
in
print shows no
to the
proportion
big changes,
but
under
granularity.
(b) When Information Theory is used for prediction:
From Figure
(13), it is
obvious
that it is very difficult to
predict
granularity
by using
Information Theory.
(c) The reason that
low
pass
SQF
filter to
film granularity is because SQF
can predict
simulate
the human
visual system and
(d) From this project we learn that when a fixed
size are used
it is better to
conditions, a
slower
4-2 Results
Compared
on
be
a
better
and a
fixed
print
under normal
words,
TRI-X 400 films
and
TRI-X
400,
T-MAX 400 film is
we used
other
flux
Theory does not.
choice.
400,
with
and
would
T-MAX
Company. In this study
quality,
film
Information
amount of photon
fast film. In the
use a
uses a
both SQF
and
a newer product
Information
from Eastman Kodak
Theory to
predict
image
films.
to study the differences between these two
(a) Under constant flux condition:
The scanning
was
done from the final
measurement contain
and scanner used
for
a systems
measurement.
by using the data in table XIII,
or without a
level
filter for
prints
of
in this
project.
MTF, including MTF
According to
the
when under constant
measurement
Therefore this
results
flux it is
of
film,
paper, lens
from this
clear
project and
that
the granularity of T-MAX 400 film
53
either with
is
smaller
than TRI-X 400
compared to
film. In the
other
words, the T-MAX 400 film is
a
finer film
when
TRI-X 400 film.
(b) Under normal exposure condition:
By using the
without
using
compared to
granularity
in the
used
same
a
Table XIII,
filter it is
of
result of granularity
T-MAX 400 film has finer grain,
8.0 Vs 9.3 for T-MAX 400. It is
system or
can make a
fixed
obvious that the
exposure, the
TRI-X 400 film. When the filter is used, TRI-X 400 film has
the truncation of data
photographic paper used
(c) We
under normal
is
possible that this
during the
scanning
is
caused
process.
when
a
by the filter
Possibly the
a cause.
brief conclusion for this
amount of photons were used
in
a
project as:
When the
system, it is better to
54
print size
use a
is fixed
fast film.
and a
V REFERENCES
James A.
Master
Wisner,
Film
Thesis, R.I.T.,
Granularity and the Effect on
Subjective Image Quality,
1986
Lisson, G, Digital Image Modeling of Film Granularity
Pictorial
Quality,
Granger,
E. M.
with
J.C.
Master
Thesis, R.I.T.,
and
Merit Function
Subjective Image Judgments, Photogr. Sci. Eng. 16
Jones, R. C.
Information capacity
Levi, L. On the
photographic recordings.
Altman, J. H.
and
dynamic
J. Opt. Soc. Amer.
Zweig, H. J. Effect
content of photographic recordings.
(SQF),
,221
Subjective
,
of spread
,
9
California,
information
,(1958
function
on
Photogr. Sci. Eng. 7
55
)
1974.
films. J. OPSoc.Amer.,51.1159.
range and
48
which correlates
(1972
10 Academic Press Inc.
of photographic
effect of granularity on
on
1983
&, Cupery, K. N. An Optical
Dainty, & R. Shaw. Image Science
Effect
,
(1961)
content of
)
the
storage of information
173, (1963 ).
W.S.
Togerson, "Theory
and
Method
of
Scaling"
John
Wily
and
Sons,
New
York,
1958.
G. P
M. J.
Clayton,
Photogr. Sci. Eng.
(1982)
M. C.
Corey,
Davidson,
and
K. N. Cupery. Scene Dependence
J. Opt. Soc. Am.
58,
1300 (1968).
56
of Image
Quality
VI APPENDICES
APPENDIX
The followings
1
-
are subjective ratings of 21 photographs
from 33 different
viewers.
Data:
The
Gray Card:
abbreviations are as
G
follow:
Egg:E
Flower: F
T-MAX100/25mm: 1
T-MAX 400 / 25
TRI-X 400 / 25
TRI-X 400 / 50
mm:
T-MAX 3200 / 135
4
mm:
mm:
mm:
5
2
T-MAX 3200 / 25
T-MAX 400 / 50
mm:
mm:
3
6
7
Gl G2 G3 G4 G5 G6 G7 El E2 E3 E4 E5 E6 E7 Fl F2 F3 F4 F5 F6 F7
Viewer #1
234422234543223455331
Viewer #2
344432243342324444342
Viewer #3
23
Viewer #4
345343234543313454332
Viewer#5
34534
1143542222444332
Viewer #6
34433
1134553122344231
Viewer #7
333411155554214555231
Viewer #8
2232
Viewer #9
23
3432344443334444342
12
763
2
122332112342
146752313
57
121
677421
Viewer #10
467642
155775214775
332
Viewer #11
2
234322223
342
Viewer #12
466644356565323
776673
Viewer #13
4
677
Viewer #14
54763
1257643125
576321
Viewer #15
45
763
2
145363213
576421
Viewer #16
44
5
422235542324
656433
Viewer #17
3
4643
3
255654324456332
Viewer #18
5
6
4
256665545
Viewer #19
44653
Viewer #20
5
Viewer #21
437422145762212363243
Viewer #22
6
Viewer #23
455432246754214567221
Viewer #24
345521124552124454342
Viewer #25
3
Viewer #26
345533234543223454342
Viewer #27
456642
144654223565421
Viewer #28
456443
145655114565432
Viewer #29
445
Viewer #30
555432353443224455332
Viewer #31
3
Viewer #32
336532225662222344223
Viewer #33
446533245654323455332
2
4
2
7
5
2
6
4
4
3
4
2
3
366775324
7
5
44
44423
4
6
53
5
2
666
533
443
1146764314565241
66644256665535
6
334
3
2
366674435
244554333
135553223
344654225
58
565442
577443
354322
344232
566
542
APPENDIX
The
following Line
Spread Function data was
2
obtained
by using Photoshop 2-5.
(a)T-MAX100/25mm:
Pixel #
Edge Reflectance
Line Spread Function
1
231
2
230
1
3
228
2
4
226
2
5
221
5
6
213
8
7
200
13
8
184
16
9
162
22
10
144
18
11
130
14
12
124
6
13
119
5
14
117
2
15
114
3
16
111
3
17
109
2
59
(b)
T-MAX 400 /25
Pixel #
1
2
mm:
Edge Reflectance
Line
Sp
227
226
1
3
226
0
4
223
3
5
220
3
6
215
5
7
209
6
8
200
9
9
187
13
10
172
15
11
154
18
12
137
17
13
128
9
14
121
7
15
116
5
16
113
3
17
111
2
18
108
3
19
107
1
20
104
3
60
(c) T-MAX 3200 /25
Pixel #
mm:
Edge Reflectance
Line
Sp
1
225
2
224
1
3
223
1
4
221
2
5
218
3
6
213
5
7
206
6
8
196
10
9
180
16
10
159
21
11
137
22
12
121
16
13
110
11
14
103
7
15
98
5
16
96
2
17
92
4
18
88
4
19
85
3
20
82
3
21
81
1
61
(d)
TRI-X 400 /25
Pixel #
mm:
Edge Reflectance
Line Spi
1
225
2
221
4
3
218
3
4
215
3
5
207
8
6
194
13
7
178
16
8
158
20
9
139
19
10
129
10
11
119
10
12
116
3
13
109
7
14
104
5
15
100
4
16
99
1
62
(e)
TRI-X 400 / 50
Pixel #
mm
Edge Reflectance
Line Spi
1
223
2
222
1
3
220
2
4
218
2
5
212
6
6
194
18
7
162
32
8
129
33
9
109
20
10
101
8
11
97
4
12
95
2
13
93
2
14
91
2
63
(f) T-MAX 400 /50
Pixel #
mm:
Edge Reflectance
1
229
2
227
3
Line
Sp
2
225
2
4
215
10
5
186
29
6
141
45
7
119
22
8
112
7
9
104
8
10
97
7
11
97
0
12
96
1
13
94
2
64
(g) T-MAX 3200 /135
ixel #
1
mm
Edge Reflectance
Line Spi
232
2
233
-1
3
222
11
4
150
72
5
89
61
6
88
1
65
APPENDIX
Written Mathcad 4.0
program and results
o.. 11
=
x
.=
0
x,4
=
2.3S
x,5
=
2.3S
=
1.59
x,
=
x,
=
1.59
x.fi
x.
=
1.59
x,_
=
1J9
x_.
=
3.97
x,.
-
1.59
x5
x5
x.
.79
1
=
6.35
=
10.32
x:o
=
12.7
*:i
--
I7-6
x.,
=
14.29
X.,
1
^
x,y
x10
0
x,?
=
=0
=
=
=
=
0
0
11.11
x..
x,,
=4.76
x.,
=
3
for Information MTFs
(a)T-MAX100/25mm:
1
-
=0
3.97
66
=
c
N
j
CFFT(x)
=
last(c) N
=
71
:=0..N
1.389
1.3S9
1.334
0.017
0.96
0.79
0.033
0.852
1.59
0.051
1.184
0.984
0.70J
1.59
0.566
3.97
0.457
6.35
0.385
10.32
0.787
0.635
0.534
0.462
0.333
12.7
0.2S5
17.46
0.396
0.327
0.235
14.29
0.184
11.11
0.137
4.76
0.098
3.97
0.071
1.59
0.055
2.38
0.045
2.38
0.04
1.59
0.037
1.59
0.03:
1.59
0.256
0.19
0.13:
-----
"
0.012
0
0.024
0
0.037
0
0.098
0.076
0.063
0.055
0.052
0.044
0.031
0.032
0.043
0.057
0.051
0.033
0.017
0.022
0.02S
0.028
0.029
0.036
0.045
0.056
0.072
0.09
0.105
0.11
0.105
0.09
0.072
0.056
0.045
0.036
0.029
Q.Q28
0.028
0.022
1
0.023
0.023
\
0.035
0.041
hh
0.037
1
1
1
/
\
1
\
/
0.024
V
o.oi:
J
0.02
0.02
0.021
0.026
0.033
0.041
0.052
0.065
0.075
0.079
0.075
0.065
0.052
0.041
0.033
0.026
0.021
0.02
L^>
50
0.016
0.02
^
67
100
(b) T-MAX400
i
&.. n
=
1
2.44
X,
=
>S
=
X.
=
4.07
X4
=
4.88
X5
=
7.32
X.
=
10.57
x.
=
12.2
1
x,
x]0
x,.
1
:=
=
:=
xu
2.44
14.63
13.32_
xu
:=
2.44
XH
:=
1.63
X15
:=
2.44
X16
:=
0.81
xn
:=
2.44
x;s
=
0
X19
:=
0
X20
:=
0
X2!
!=
0
X22: =
0
X23''
=
0
X24- =
0
7.32
=
5.69
=
4.07
I
mm:
0
X :=
Xg
/ 25
68
c :=
N
j
=
CFFT(x)
last(c)
N-71
=0..N
id
1 ji
hi
1.378
1.378
1.318
1.156
0
0.078
0.957
2.44
0.063
0.839
2.44
0.678
4.07
0.515
4.88
0.381
7.32
0.288
10.57
0.935
0.709
0.524
0.396
0.314
0.261
0.226
0.195
0.157
0.115
0.083
0.071
0.062
0.051
0.053
0.066
0.07
0.062
0.053
0.052
0.055
0.063
0.078
0.09
0.087
0.061
0.022
0.027
0.051
0.049
0.03
0.044
0.077
0.09
0.077
0.044
0.03
0.049
0.051
0.027
0.022
0.061
0.087
0.09
i
1
0.22S
12.2
0.189
14.63
0.164
13.82
0.142
7.32
0.114
5.69
0.0S3
4.07
0.06
2.44
0.051
1.63
0.045
2.44
0.037
0.81
0.033
2.44
0.048
0
0.051
0
0.045
0
0.039
0
0.038
0
0.04
0
0.046
0
0.056
0
0.066
0
0.063
0
0.044
0
0.016
0
0.02
0
0.037
0
0.036
0
0.022
0
0.032
0
0.056
0
0.06<5
0
0.056
0
0.032
0
0.022
0
0.036
0
0.037
0
0.02
0
0.016
0
0.044
0
0.063
0
r\ axx
A
----
0.055
-
0.056
0
0.046
0
0.04
0
20
10
V,
50
i
69
100
(c) T-MAX3200 / 25
=
i
o.. n
x
:=
4.64
xM
:=
3.31
x]5
:=
1.32
0
=
x
mm:
x. :=
0.66
x,
:=
0.66
x
x3
=
1.32
xn
:=
2.64
x4
=
1.99
xlg
:=
1.99
X5
=
3-31
x19
:=
1.99
Xo
=
3-97
X20
:=
66
6.62
x2]
:=
0.66
:=
10.6
x,2
:=
13.91
Xj3
x.
xg
x9
x10
-
:=
.=
16
.=
'Si
=
10-6
X12
=
7'28
1.99
:=
1.32
=
132
14.57
^4
2.64
70
c
=
CFFT(x)
N.=
j
:=
last(c) N-71
0..N
1.389
1.389
1.203
1.004
0.781
0.571
0.41
0.312
0.27
0.965
3.17
0.095
0.866
2.33
0.087
0.723
2.38
0.562
6.35
0.412
10.32
0.29::
12.7
0.224
15.87
0.194
15.08
0.187
7.94
0.186
7.94
0.13
2.38
0.26
0.258
0.251
0.226
0.163
5.56
0.131
3.97
0.181
0.121
0.087
3.17
0.041
0.79
0.057
0.01;
0.011
0.042
0.03 :
0.056
0.041
0.053
0.038
0.047
0.034
0.054
0.039
0.071
0.051
0.087
0.062
0.095
0.098
0.093
0.082
0.063
0.063
0.077
0.1
0.118
0.122
0.113
0.097
0.08S
0.097
0.113
0.122
0.118
0.1
0.077
0.063
0.068
0.032
O 001
-
0.098
1.34
0.069
0.07
0.067
0.059
0.049
0.045
0.056
0.072
0.085
0.038
0.081
0.07
0.064
0.07
0.081
0.083
0.085
0.072
0.056
0.045
0.049
0.059
71
0.07
0
0.069
0
0.062
0
(d)
TRI-X400 / 25
i
0..71
:=
x
mm:
x13
:=
3.97
:=
3.17
0
.=
i
xM
xT
:=
3.17
x15
:=
2.38
xis
=
2.38
xn
6.35
X18
x,
x,
x.
4
x5
x,0
x_
-
=0
X19
0
=
=
=
12.7
^o^0
:=
15.87
X21
=
15.03
X22
7.94
**
=
x8
x9
10.32
=
0.79
=
:=
xio
xu
x12
=
7'94
h*
=
2.38
:=
5.56
"
=
=
-
72
c :=
CFFT(x)
N.=
j
-
last(c)
N=
71
0..N
hi
1.3S8
1.333
1
1.04
0.738
0.639
0.56
0.472
0.927
0.04
0.749
0.032
0.568
1.32
0.461
1.99
0.404
3 31
0.34
3.97
0.266
6.62
0.369
0.29
0.209
10.6
0.189
13.91
0.13
14.57
0.262
0.25
0.211
0.133
0.152
10.6
0.099
7.23
0.042
4.64
0.032
3.31
0.059
0.044
0.079
0.089
0.073
0.046
0.033
0.031
0.025
0.024
0.032
0.04
0.042
0.037
0.032
0.03
0.03
0.037
0.04
0.028
0.021
0.034
0.034
0.023
0.034
0.034
0.021
0.028
0.04
0.037
0.03
0.03
0.032
0.037
-
0.042
1.287
0.057
1.32
0.064
2.64
0.052
2.64
0.033
1.99
0.023
1.99
0.022
0.66
0.03
0
0.029
0
0.023
0
1
\
i
1
1
1
0.018
0.66
0.017
1.99
0.023
1.32
0.029
1.32
\
Uv
j
0.023
0.022
5
0.027
0.029
0.02
0.015
0.025
0.025
0.02
0.025
0.025
0.015
0.02
0.029
0.027
0.022
0.023
n mi
A
w
50
0.03
0.022
J
\
0.027
0.022
!
\
73
100
(e) TRI-X400 /
50
mm:
=0.-71
i
x,3
:=
XH
=
1.52
=0
x.
i
0.70
=
x,s:S0
Xj
i---
-
xifi
X-
=
i-32
*
xn
=
x
i0
=
0
4.55
-f
xis
x:
TO
x;
-
=
.=
24.24
25.0
xie
=
^o
=0
X,.
ii
=
X?
=
15.15
Xv
=
<3.0<3
*to
=
:=
0
3.03
X22
xn
=
I3\04
!=
i.52
x,,
=
0
4rJ
x1;
=
1.52
X24
:=
74
=
c
N
j
=
=
CFFT(x)
last(c)
N
71
0..N
I
ji
1.389
1.389
U.VZJ.
I
0.033
1.369
1.311
1.222
0.986
0.76
0.944
1.52
0.88
1.52
0.8Q2
4.55
0.024
0.036
0.026
0.043
0.031
1.114
0.998
0.836
0.735
0.699
0.627
0.566
0.511
0.457
0.4
0.342
0.232
0.225
0.175
0.135
0.106
0.035
0.069
0.055
0.043
0.036
0.033
0.03
0 025
0.013
0.012
0.009
0.011
0.011
0.01
0.007
0.004
0.719
13.64
0.633
24.24
0.565
25
0.503
15.15
0.452
6.06
0.408
3.03
0.368
1.52
0.329
1.52
0.283
1.52
0.246
0.203
0.162
cjh
0.126
0.097
0.076
-kl
50
0.061
J
0.05
0.039
0.031
0.026
0.024
0.022
0.Q1S
0.013
0.008
0.007
0.003
0.008
100
0.007
0.005
0.003
-5
9.999-10
0.004
0.007
0.01
0.011
0.011
0.009
0.012
0.018
0.025
0 01
1 00
0.003
0.005
0.007
0.008
0.008
O.QQ7
0008
0.013
0.013
75
(f)
T-MAX400 / 50
0.. 71
=
i
x
X13
=
0
X14
=
0
1.48
X15
=
0
-
1.43
X,
=
0
.=
7.41
X17
=
0
=
21.48
X1S
=
0
=
33.33
=
0
=
0
=
0
=
0
=
0
=
0
0
-
i
x^
=
x,
x.
x4
x.
mm:
5
.
X19
x,il
=
16.3
*30
x.
=
5.19
V.
s
193
X22
=
5.19
X23
*
x?
'-"
xio
x
x,,
X.
=
=
0.74
1.43
76
=
c
N
j
CFFT(x)
=
last(c) N
=
71
=0..N
!
Ji
1.389
1.389
1.387
1.381
1.37
0.999
0.994
0.481
7.59
49.66
0.987
42.07
0.976
0.69
1.356
1.338
0.53
0.579
0.963
1.315
0.947
1.29
0.923
1.26
0.9Q7
1.228
0.884
1.192
0.353
1.154
0.331
1.113
0.302
1.07
0.771
1.025
0.979
0.931
0.332
0.832
0.781
0.731
0.68
0.63
0.579
0.53
0 481
0.433
0.386
0.34
0.295
0.251
0.207
0.165
0.123
0.032
0.042
0.01
0.042
0.032
0.123
0.165
0.207
0.251
0.295
0.34
0.738
0.705
0.67
0.635
0.599
0.563
0.526
0.49
0.453
0.417
0.332
0.346
0.312
0.278
0.245
0.212
0.181
0.149
0.119
0.039
0.059
0.03
0.007
0.03
0.059
0.089
0-119
0.149
0.181
0.212
0.245
77
0.346
0
0.382
0
0.417
0
(g)T-MAX3200/135mm:
i
0..71
=
xu
X14
x^
=
7.59
x,5
x,
=
49.66
xid
x,
=
42.07
xn
x.
=
0.69
x_
2
4
X19
=
0
u
x.
x,
=
0
x:i
=
0
x
=
0
X23
\
X9
X10
xu
X12
=0
:=
=
=
0
=
0
=
0
-
.
20
0
=
0
=
0
x.
0
=
=
x
x5
0
0
x :=
X5
-
0
=
0
=
0
=
=
0
78
c
N
j
CFFT(x)
=
:=
=
last(c)
N
=
71
0..N
c!
1.389
1.389
1
1.371
1.43
0.949
1.43
1.313
1.238
0.891
1.14
1.034
0.931
21.48
0.744
33.33
0.67
16.3
0.603
0.756
0.223
0.09
0
0.124
0
0.16
0
5.19
0.544
5.93
0.494
5.19
0.625
0.172
7.41
0.32
0.837
0.686
-
0.125
0.987
0.4::
0.57
0.411
0.74
0.376
1.48
0.522
0.432
0.347
0.451
0.431
0.31
0.419
0.411
0.401
0.3 S4
1
0.325
\
0.302
1
"\
0.296
0.289
0.257
50
0.23
J
0.197
0.223
0.172
0.16
0.124
0.125
0.09
0.087
0.063
0.061
0.043
0.048
0.057
0.044
0.034
0.035
*"\
0.041
0.069
0079
0.087
0.091
0.092
0.093
0.092
0.091
0.087
0.079
0.069
0.057
0.048
0.048
0.057
0.062
0.065
0.066
0.067
0.066
0.065
0.062
0.057
0.049
0.041
0.035
\n
y
\^_~j\
0.357
0.274
1
V
0.277
0.32
i
1
mjl
79
20
4
APPENDIX
Written Mathcad 4.0
program and results
-
4
for SQFs
(a)T-MAX100/25mm:
i
=
<).. 8
j:=0..4
i -.25
vx. : =
i
vy0:=l
vyr=. 96
vy2=. 852
vy3:=. 708
vy4:=.
566
vy3:=.
?57
^6--
385
vy7-=333
vy8=.:285
fre%
freqj
freq2
freq3
freq4
modi^
=
.5
=
.707
=
1
=
1.414
=
2
)
:
=
linteiWvx
vy.freq.)
mod(j)
0.852
0.733
0.566
0.41
0.285
.
.5-(mod(0)
sqf :
+
mod(4)) + mod( 1
) + mod(2)
+
4
sqf
=
0.569
80
mod(3)
(b) T-MAX400 / 25
i: = 0..8
mm:
j: = 0..4
vx, :=i-.25
vy0:=i
vyr=.957
vy2:=.839
vy. :=.678
vy4=.515
vy5:=.381
vy
.=
.288
vy?:=.228
vv
:=.189
freV=.5
:=
freqi
freq2 1
freq3 -=1.414
.707
:=
freq4:=2
mod(j)
=linterp/'vx,vy,frea
modCj)
0.839
0.706
0.515
0.32
0.189
.5-(mod(0)
sqf
+
mod(4))
+
mod( I
)
+
mcxl(2)
4
sqf
=
0.514
81
~
mod(3)
(c) T-MAX3200 / 25
i: = 0..8
mm:
j: = 0..4
vx. :=i-.25
i
vy0=i
vyr=.927
vy2:=.749
vy3=.568
vy4:=.461
vy5
:=.404
vy6=.34
vy?
=.266
vyg:=.209
frev=.5
freq1=.707
freq2
=1
freq
:=
1.414
freq4=2
mod(j)
:=linterp/vx
vy.freaj
mod(j)
0.749
0.599
0.461
0.362
0.209
.
sqi
:=
.5-(mod(0)
+
mod(4))
+ mod(l
) + mod(2)
4-
4
sqf
=
0.475
82
mod(3)
(d) TRI-X400 /
i: = 0
25
mm:
j:=0..4
-8
i-.25
vx. : =
1
1
vy0:=
vyr=
.965
=
vy2
.866
vy3: =
.723
vy4:=
.562
vy5: =
.412
vy6
=
vy7
=
224
vyg
=
194
.295
fre%
freqi
freq2 :=1
=
=
freq.
.=
.5
.707
1.414
freq4:=2
mod(j )
=
linterp (vx
,
vy frea
,
mod(j)
0.866
0.748
0.562
0.335
0.194
+
sqf
sqf
'=
=
.5-(mod(0)
mod(4))
+
mod( 1
) + mod(2)
4-
0.544
83
mod(3)
(e) TRI-X400
i: = 0..8
vx.
/ 50
mm:
j:=0..4
-1-.25
i
vy0
=1
vyj
=.986
vy2
=.944
vy3=.88
vy4=.802
vy5:=.719
vyg
=.638
vy?:=.565
vyg
=.503
freq():=.5
freq
"=.707
freq2
freq3
:=
.=
1
1.414
freq4: = 2
mod(j)
=linterpi
vx,vy,frea)
mod( j )
0.944
0.891
0.802
0.666
0.503
4-
sqf
sqf
_
=
.5-(mod(0)
mod(4))
+
mod( 1
)+
mod(2)
4-
mod(3)
0.771
84
(f) T-MAX400 / 50
i: = 0
vx.
vy0
.=
=
vyi: =
vy2: =
vy3:=
vy4
j: = 0..4
8
i-.25
1
.987
.949
.891
=
.82
vy5.=
vy6
mm:
.744
67
=
vy/=
603
vyg:=
544
freq0 :=.5
freqj
freq2 1
freq3 1.414
=
.707
=
=
freq.
=
mod(j)
2
=linterp/'vx,vy)freq.)
mod(j)
0.949
0.901
0.82
0.695
0.544
4-
sqf:
.5-(mcxl(0)
mod(4))
4-
mod( 1
)
4-
mod(2)
sqf =0.791
85
4-
mod(3)
(g) T-MAX3200 /
i=0..8
135
mm:
j:=0..4
vx. :=i-.25
vy0=i
vy :=.999
vy2:=.994
vy, :=.987
vy4:=.976
vy5
:=.963
vyg:=.947
vy7:=.928
=.907
vy
te%
=.5
freqr=.707
1
freq2
freq3
freq4
=
1.414
2
mod(j)
=
linterp/'vx,vy,frea
mod(j)
0.994
0.988
0.976
0.953
0.907
4-
f
._
.5-(mod(0)
mod(4))
+
mod( 1
) + mod(2)
4
sqf
=
0.967
86
4-
mod(3)
APPENDLX
Written Mathcad 4.0
(a)
T-MAX100 / 25
mod
mod,,
:
=
=0
:=0
mod,.,
XL
.852
mod,,
=0
jo
=
mod
mod,4
Capacity results
jl
=.96
mod
Information
mm:
1
:=
mod
program and
5
-
.708
.=
.566
mod
:
=
mod,
:
=
mod
:
=
.457
mod,,
34
:=0
mod,,
=0
mod,,
=0
Jo
noise
=
0
C
mod9
noise
=
.285
=
.235
mod1Q
=
.184
modu: =
.137
mod12: =
.098
mod13: =
.071
mod14
=
.055
mod15-
mod16
modn
modlg
mod19
mod2()
mod21
mod22
m0d23
mod24
mod25
mod26
mod27
mod2g
mod29
mod30
C
1.0-
=
.333
i
modg
(mod.)'
36
.0325
.385
=
.045
=
.04
=
.037
=
.032
=
.023
=
0
=
0
=
0
=
0
=
0
=
0
:=0
:=
0
=
0
:=
0
:=0
87
=
0
=30.85
(b) T-MAX400 / 25
mod0
:=
modx
:
1
=
mm:
=
0
mod32: =
0
:
mod
.957
mod2:=.839
mod33
=0
mod,,
34
=0
=0
mod3
:=.678
mod_,
:=.515
mod
mod3
:=.381
mod,,
36
mod, :=.288
O
mod7:=.228
noise
:
=
0
36
=.0334
mod.
C
:
2
=
i
mod.
.189
:=.164
modg
1 42
mod10
:
mod^
:=.H4
=
.
mod12:=.083
mod13
:=.06
mod,,
14
:=.051
mod15
:=.045
mcxi
:=.037
16
modn=.038
modlg=.048
modig:=.051
mod2():=.045
mod21
:=.039
mod^
=.038
mod
=0
mod,,
24
:=0
mod
=0
mod26:
=
0
:=
mod
27
0
mod2g:=0
mod^-0
mod30:=0
88
=
0
^g
C
1-0
noise
=29.991
(c)T-MAX3200/25mm:
mod
:=
mod1
:
1
=
mod2:=.749
:=.568
mod3
mod, :=.461
mod
=0
m0d31
.927
:=.404
mOd32:=0
mod33:=0
mod,,
34
=0
0
mod35
:
mcxi
:=0
=
JO
mod,
.
=
.34
6
mod
l7
=.266
noise
36
=.0486
C
1.0-
i
modg:=.209
mod
-9-.189
:=.
mod1():=.18
modn=.152
mod12:=.099
mod13
:=.042
mod,,
14
:=.032
mod15
:=.057
mod,, :=.064
16
mod17:=.052
mod,:=.033
mod19:=.023
mod2():=.022
mod21
:=.018
mod^-,017
mod^
:=.023
mod,,
24
:=.029
mod^-,03
mod,,
:=.027
mod27:=.023
mod2g
:=
.022
mod^,:=.022
29
mod3():=0
89
C
noise
=
0
=24.818
(d) TRI-X400 /
1
mod
:=
mod
:=.965
mod
:
mod
mod,
mm:
mod31: =
=
0
mod32:=0
mod33
:=0
:=.723
mod^
34
:=0
=.562
mod35
:
412
mod,,
Jo
:=.295
noise :
mod,
mod3
25
.866
:=
mod7:=.224
=
0
=0
36
=
.0303
mod.1)
C
log
=
1.0-
C
noise
modg:=.194
i
=
0
mod9:=.187
mod1():=.186
mod
:=.18
mod
:=
.163
mod:=.131
Ij
mod,,
14
:=.087
mod15:=.041
mod,,
16
:=
.011
mod17:=.03
modlg:=0
mod19=0
mod20:=0
mod
mod
:
=
0
=0
0
mod
:=
mod,,
24
:=0
mod
:=0
m0d26: =
mod
:=0
mod28:=0
mod^^O
mod30:=0
90
=31.148
(e) TRI-X400 / 50
1
modQ:==
:==
mod
=
mod
=
mod,4
=
mod
=
mod,0
=
mod
mod3I
.986
mod
:=0
mod32=0
.944
mod33
:=0
.88
mod,,
34
:=
.802
mod,,
j5
=0
.719
mod,,
i6
:=0
638
=
mm:
.565
noise
0
36
=.0303
c
1.0-
[MF
noise
modg.
=
:
=
mod
i
=
.503
.452
9
mod10
:=.408
modn
:=.368
mod12
:=.329
modu
=
modM
=
mod15
=
mod16
=
.288
=
modn
modlg: =
mod19:
=
mod20:
=
mod2i
=
.246
.203
.162
.126
.097
.076
m0d22: =
.061
.05
.039
m0d23'
=
mod24
:
==
m0d23::=
m0d26::=
mod
mod
mod
'=
.031
.026
.024
.022
.018
27
:=
.013
28
:=
.008
29
mod,
:=
.007
30
91
0
J
C
=47.06
(f) T-MAX400 / 50 mm:
:=1
mod0
mod
mod1:=.987
949
:=
mod
:=0
mod32:=0
mod,, :=0
j3
mod3
:=.891
mod,
=.82
mod5
:=.744
mod,,
mod,
:=.67
noise :=
mod,,
:=0
34
mod
=0
:
=
0
36
.0331
]T
mod7:=603
i
:
=
.494
mod
:=.45
mod
:=.411
mod
:=.376
mod
:=.347
mod,,: =.325
14
:=
mod
.31
mod:=.302
16
modn:=.296
mod,,
=.289
mod19:=.277
mod
:=.257
mod
:=.23
mod^^.197
mod
mod,,
24
mod^
mod,,
mod
C
lodl.Onoise
mod, :=.544
mod
(mod.)-
f
:=.16
=.124
:=.09
=.063
.=
.044
mod:=.034
mod^^O
mod30:=0
92
=
0
=63.175
(g)T-MAX3200/135mm:
i: = 0..36
1
:=
mod
mod
:=.999
mod
:=.994
mod3i
=
mod,2
=
mod,,
=
mod,
=
mod35
=
mod36
=
.149
.119
.089
3j
mod3
.=
.987
.059
o4
mod,4
.=
.976
mod5
:=.963
mod,6
=.947
mod7:=.928
noise
=
.03
.007
.0621
(mod.
c
i
:=.907
modg
mod9:=.884
mod10
:=.858
mod
:=.831
mod12=.802
mod13
:=.771
mod,,
14
:=.738
mod13
:=.705
mod,,
16
:=.67
mod
=.635
modTO:=.599
lo
mod
=.563
mod, :=.526
20
:=.49
mod
mod
=.453
mod
:=.417
mod,,
24
:=.382
mod
.=
.346
mod,,
26
.
=
.312
mod27:=.278
mod
.
=
2>
=
.245
25
mod2g:=.212
mod30:=.181
93
=
0
1.0
C
-r
noise
=63.545