See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/258720999
Application of LiDAR and optical data for oil palm plantation management in
Malaysia
Article in Proceedings of SPIE - The International Society for Optical Engineering · November 2012
DOI: 10.1117/12.979631
CITATIONS
READS
15
2,656
5 authors, including:
Helmi Shafri
Mohd Hasmadi Ismail
Universiti Putra Malaysia
Universiti Putra Malaysia
243 PUBLICATIONS 4,174 CITATIONS
131 PUBLICATIONS 1,101 CITATIONS
SEE PROFILE
Mohamad Khairil mohamad Razi
3 PUBLICATIONS 21 CITATIONS
SEE PROFILE
Izzuddin Anuar
Malaysian Palm Oil Board
23 PUBLICATIONS 272 CITATIONS
SEE PROFILE
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Socio-ecological modelling of the consumptive and non-consumptive uses of wetlands in Peninsular Malaysia View project
BSR survey in oil palms View project
All content following this page was uploaded by Izzuddin Anuar on 23 May 2014.
The user has requested enhancement of the downloaded file.
Application of LiDAR and optical data for oil palm plantation
management in Malaysia
Helmi Z. M. Shafria, Mohd Hasmadi Ismailb, Mohd Khairil Mohd Razic, Mohd Izzuddin Anuara and
Abdul Rahman Ahmadd
a
Department of Civil Engineering, Faculty of Engineering, Universiti Putra Malaysia (UPM); b
Faculty of Forestry, UPM;c Felda Technoplant Sdn Bhd, Tingkat 3, Anjung Felda, 53000, Kuala
Lumpur.; dFederal Land Development Authority( FELDA), Bahagian Pertanian, Tingkat 2, Wisma
Felda,53000, Kuala Lumpur.
ABSTRACT
Proper oil palm plantation management is crucial for Malaysia as the country depends heavily on palm oil as a major
source of national income. Precision agriculture is considered as one of the approaches that can be adopted to improve
plantation practices for plantation managers such as the government-owned FELDA. However, currently the
implementation of precision agriculture based on remote sensing and GIS is still lacking. This study explores the
potential of the use of LiDAR and optical remote sensing data for plantation road and terrain planning for planting
purposes. Traditional approaches use land surveying techniques that are time consuming and costly for vast plantation
areas. The first ever airborne LiDAR and multispectral survey for oil palm plantation was carried out in early 2012 to
test its feasibility. Preliminary results show the efficiency of such technology in demanding engineering and agricultural
requirements of oil palm plantation. The most significant advantage of the approach is that it allows plantation managers
to accurately plan the plantation road and determine the planting positions of new oil palm seedlings. Furthermore, this
creates for the first time, digital database of oil palm estate and the airborne imagery can also be used for related
activities such as oil palm tree inventory and detection of palm diseases. This work serves as the pioneer towards a more
frequent application of LiDAR and multispectral data for oil palm plantation in Malaysia.
Keywords: LiDAR, oil palm, plantation, precision agriculture, multispectral, airborne, planting, road planning
1. INTRODUCTION
Malaysia oil palm industry is the leading commodities and one of the major contributors to the Malaysia’s economy after
oil and gas sector. Malaysia and Indonesia are the major commodity producers with Malaysia currently being the world’s
second-largest area of oil palm after Indonesia. Together these two countries account about 84% of total world
production and 88% of global exports. With the increasing price and demands for the Crude Palm Oil (CPO) and with
the 4.69 million hectares that were planted with oil palm trees, plantation industry and estate managers have to look into
the most crucial factor that will decide the yield and outcome from each of its decision in estate management [1].
Due to the importance of oil palm to the country, accurate and reliable information is needed for oil palm plantation
management, not just on plant quality, but also on phenology, health and yield prediction. Infrastructure can deliver
major benefits in economic development and reduction of waste and environmental sustainability [2]. Improvement of
the productivity and quality of oil palm in operation is one of the toughest challenges for the oil palm mangers And
before it goes too far on the phenology, health and yield prediction, the most important and crucial is the basic earthwork
itself, the replanting activities which involved the building of the basic plantation infrastructure such as agricultural road
and planting terraces are the most important factor in plantation management. The second replanting is the chance to
manage, to improve and to make total correction on the first replanting mistakes that has been made and thus reorganized
the structure of an estate that will leads to a high productivity yield production since the estates logistics and
infrastructure already improved. Terraces and road that well constructed in replanting area deals with all aspects of
everyday works of a well manage estates. A well and sound constructed terraces and agricultural road enable make
Lidar Remote Sensing for Environmental Monitoring XIII, edited by Kazuhiro Asai,
Nobuo Sugimoto, Upendra N. Singh, Achuthan Jayaraman, Jianping Huang, Detlef Mueller,
Proc. of SPIE Vol. 8526, 852608 · © 2012 SPIE · CCC code: 0277-786/12/$18 · doi: 10.1117/12.979631
Proc. of SPIE Vol. 8526 852608-1
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 11/26/2012 Terms of Use: http://spiedl.org/terms
transporting oil palm trees from nursery to planting platform a lot more easier. A well constructed terraces can perform
as a water retainer and as a planting platform [3].
In Malaysia, the Federal Land Development Authority (FELDA) was established with the objective of carrying out land
development and resettlement of the landless, it has to date developed 853,313 ha of land and resettled 112,635 families.
FELDA was established on 1 July 1956 under the Land Development Act 1956 as a result of recommendation of
Government Working Committee [1]. Originally, FELDA was functioning as a board to manage and channel financial
aids to the state Government to carry out land development schemes in the respective states besides coordinating land
development in these states which include the movement of population within the states. The major agricultural activity
of FELDA is oil palm crops Since the establishment of FELDA in 1956, there is no clear and detailed way to assist
plantation manager making accurate decision whether it involved harvesting activities, manuring, replanting and etc. It
all based on the experience of the manager and the usage of the conventional way in managing oil palm plantation and it
is not very accurate. One example is on topography aspect in replanting area. Some researchers and experts recommend
no planting above 40º and that platforms are sufficient up to 20º , but some researchers considered that 20º should be the
final upper limit. Plantation with slopes up to 30º (60%) will be having difficulties in harvesting process, and there is
always danger of erosion. Palms should not be planted on slopes of up to 30º , although it is best not to exceed 12º. On
broken land, the variation of slope makes road layout difficult, as road density needs to be high because of the difficulty
of in-field transport, therefore analysis on the slope is critical aspect in re-planting planning. Since the second and third
generations of replanting for FELDA plantation area on its way, analysis on the slope is a critical aspect in re-planting
planning [1].
Previous research on the use of remote sensing technology for oil palm assessment were conducted by using airborne
hyperspectral systems [4-6]. Despite the success of these techniques, the cost of operation is very high and requires very
complex processing skills. Furthermore, the use of airborne hyperspectral imagery could not provide the digital terrain
information needed for road planning and planting. Aerial remote sensing techniques utilizing LiDAR and optical
sensors is accepted as the most efficient and cost-effective means to create accurate topographic, digital elevation and
terrain data [7]. It has become the standard for flood mapping and many other applications requiring fast, accurate,
inexpensive Digital Terrain Models (DTMs), Digital Elevation Models (DEMs) and other geospatial features [8]. LiDAR
technology provides an important tool for the management of plantations. Airborne and ground based data called
ECHIDNA and ECHIDNA Validation Instrument (EVI) can be linked and matched through consistent models for native
and plantation forest inventory [9]. The study proved the potential for new methods of forest and plantations assessment
and the development of efficient allometry from both ground and airborne.
Data obtained in the field are difficult to obtain and in many cases, inaccurate. Typical examples include plantation
replanting boundaries varying from FELDA to its subsidiaries Felda Technoplant Sdn Bhd, and applied production areas
are different from the actual. This has been a result of the problems in measurement and mapping of difficult terrain and
remote inaccessible locations. Furthermore, plantation management has to consider the changing nature of an estate that
extends from initial land clearing, the production stage and finally the re-planting or conversion phase. LiDAR differs
from traditional methods and provides an alternative toolsto monitor and analyze data. By using LiDAR, remote and
inaccessible area can be more efficiently and effectively managed, making replanting easy. FELDA’s human resources
management is also a factor that needs to be considered in applying this technology. This is because Felda now consists
of 60% senior staff and out of those, approximately 50% are ranging from 21 to 36 years of age. This condition creates
a state of denial when the senior staffs are reluctant to accept new technology such as LiDAR.
Proc. of SPIE Vol. 8526 852608-2
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 11/26/2012 Terms of Use: http://spiedl.org/terms
The general objective of this preliminary study was to assess the feasibility of using airborne remote sensing in assisting
oil palm plantation management in Malaysia. The specific objectives of this initiative are:
a.
To introduce the usage of airborne remote sensing for assisting FELDA plantation managers in making accurate
decision with detailed land information and acquired faster hence cheaper information for decision making in
replanting area.
b.
To study the possibilities of using LiDAR and optical sensors methodology in all FELDA replanting area and
its impacts.
2. MATERIALS AND METHODS
2.1 Study Area
The study was carried out at FTPSB Ijok and FELDA Gunung Bongsu. Felda Technoplant Sdn Bhd (FTPSB) was
incorporated in 2005 as a private limited company under the plantation group fully owned by FELDA. In this study, the
specific site is at Felda Ijok located at Selama, Perak, Malaysia. Felda Ijok consists of 654 registered FELDA settlers and
3200 ha area planted with oil palm. The new replanting area is located in Phase 1 and 2 at the Felda Ijok. Felda ijok
geographical Coordinate is 5 06’ 44” N to 100° 46’ 31” E, respectively. Felda Gunung Bongsu geographical coordinate
is 5°10’23.06 N and 100° 51’ 50” E, It consisst of 154 settlers with 1500 hectare of oil palm areas. Figure 1 shows the
location of the study area in Ijok.
Felda Ijok and Gunung Bongsu are chosen because they are new replanting areas and consist of 70% of hilly areas. The
main reason to use remote sensing technology is that the terraces are designed not according to the specifications because
of the hilly area of more than 40°. Redesigning terraces and constructing new terraces with new agricultural road in this
replanting area are very challenging if conventional way such as traditional leveling is used because it cannot give
accurate data regarding the DEM and topographic aspect.
Figure 1. Location of Felda Ijok
2.2 Methods
Data acquisition was carried out using a helicopter platform (figure 2) with the standard operating principle (figure 3).
The airborne set equipment consist of LiteMapper 5600, Inetial Measurement Unit (IMU) 256Hz, in-flight data recorder
560, Computer Controlled Navigation System (CCNS), Terasolid, Microstation, GIS Sofware, GPScas, WinMP (figure 4
to 6) as the most crucial items in this study. LiDAR data was acquired by Airborne Informatics Sdn Bhd, a private
Proc. of SPIE Vol. 8526 852608-3
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 11/26/2012 Terms of Use: http://spiedl.org/terms
company specializing in aerial remote sensing data acquisition. The data capture was supported by Trimble R7 and
DigiCAM 39 Megapixel high resolution system. Data collection was assisted by five Felda Ijok and Gunung Bongsu
officers with three Airborne Informatics Sdn Bhd personnel. Collection of data involves acquiring geographical
information of the study site such as Felda Ijok and Gunung Bongsu topographical map, DEM, DTM and Orthophoto.
Before the mission, flight preparation, pre flight preparation and post flight preparation will be conducted by the person
in charge. Flying was done at 500-1000m elevation over the target area. After the mission, the data and image weree
processed and converted into GIS database. Prior to that, ground sampling needed to be done to support interpretation.
Map will then be generated depicting land structure and condition using Autocad MAP 11, Global Mapper and ERDAS
9.1. Data were stored, processed and displayed in ArcGIS 9 (ArcMap Version 9.2) referenced to the WGS84 datum with
a Universal Transverse Mercator (UTM) projection, zone 47 North. Once finished, drawing files were imported and
converted into an ArcGIS geodatabase,
LIDAR SYSTEM
Wiring
GPS ANTENNA
's
4) 10.00 m
32.81 ft dia
A
3.40 m
11.15 ft
3.08 m
10.10 ft
1r
0.56 m
1.83 ft
-
V
4
o 50 m
1.64 ft
9.60 m
31.51 ft
11.52 m
- 37.79 ft
4.92 ft
2.07 m
6.79 ft
2.60 m
8.53 ft
Figure 2. LiDAR position and mounting on helicopter platform
41-.
GPS satellites
Direction of the
flight
ln
10
40
ID
4.
41.
OP
GPS base station
Figure 3. Principle of LiDAR data acquisition operation
Proc. of SPIE Vol. 8526 852608-4
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 11/26/2012 Terms of Use: http://spiedl.org/terms
Figure 4. Global Positioning System (GPS) for ground control point collection
1
Laser scanner LMS -Q560
CCNS4
(Computer Controlled Navigation System
411) generation)
IMU (Inertial Measurement Una)
8" TFT touchscreen
LMcontrol
5" TFT display
AERQcontrol
Shock -absorbing mount
Dala recorder DR560
Figure 5. LiDAR system
Figure 6. Hasselblad 39Megapixel Camera for Orthophoto
Proc. of SPIE Vol. 8526 852608-5
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 11/26/2012 Terms of Use: http://spiedl.org/terms
3. RESULTS AND DISCUSSION
3.1 LiDAR Analysis
Figures 7 to 15 show the output of the LiDAR mapping work. There are perimeters of the area, 3D imagery, orthophoto
with road and terrace design for the plantation area. Plantation manager utilizing the LiDAR in this study found out that
LiDAR is a very much useful for them to plan and make accurate decision on new road planning and terraces
construction. It also helps them to automatically determine the quantity of the terraces and new agricultural road needed
per hectare and this make the estimating cost for infrastructure planning more accurate.
The usage of LiDAR technology also makes palm counting a lot more easier. By using LiDAR in this replanting areas,
the quantity of oil palm seedlings that need to be planted in the field is automatically known. It can help the management
to plan the quantity of the seedlings needed to be produced to plant at the field. In the mean time, data that have been
stored into the GIS system allows the management to keep track of individual palm each month and year. It contributes
to the accurate application of fertilizer which in conventional way, there is no way to determine an area for fertilization.
Previously, fertilizer will be cast manually and will be ordered each year just by assumption.
Figure 7. 3D DEM image for Felda Ijok
Proc. of SPIE Vol. 8526 852608-6
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 11/26/2012 Terms of Use: http://spiedl.org/terms
in
125 in
25U in
375 in
Lin
1542 UTM ( 1,,G.3,) -(E,190.911, 572938.,;4 ) 51[51.4"N, 1004 44* 34.
Figure 8. DEM Image for Ijok
Figure 9. 3D DEM image for Ijok
Proc. of SPIE Vol. 8526 852608-7
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 11/26/2012 Terms of Use: http://spiedl.org/terms
.,
.
' ';4'1,;'IT:Tki0147;71r;tt
..-Ttaft-Iti
' kt, itt A ill',0110 r
Tettittl 4,TI
tt lit
Atr1
kaiN
i Vitt' NTT
t
\,
rat alt
tr':;f
- , ii tf
titlial;11.1;11:1StZfIL
tt tijkii.'7::'
It 'ft 'ITTATT;.t
..ft:','?":11At
.4,,,,,
T:
,
-ftv,:,
m mi
'
' .0, ar",i*.
t
l't 44^
k;,
, .'
ST :
,.' '
-.,
.
.11%
AA kaN
.."
t
"
It- -
"
47
.,
-,
.
-Thtt Aft'
44,0ktii
'
t ' m ,i,,1
a "Al..3%, ''
,.rt :i I. .
'i%r
..
t attti1It 4tt III 'tit/
Figure 10. Planting point design for Gunung Bongsu
N
1 T 4, 'tie 1- ' '1
e.; l';, eAS;,t'N'Tfi, ,1 -"1,\Ill '',". 4 A.,f,t''"4,10'/' .
.
'1r
'
mi, N 0,1,7.
',t4t1;
.;.
"M; 114{" Alt" ,te " " ' "QIN:r.'A.47:,;k .11",:m
r:"::. 4 m';Attt11:11'.
"Y"ttik.11-T
-
iti,1,,fi...;;T:,
m
:-A
Figure 11. Before planting point was inserted
Proc. of SPIE Vol. 8526 852608-8
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 11/26/2012 Terms of Use: http://spiedl.org/terms
t:r 4^,
A
t
1
3
Figure 12. Contours generated from Laser Point
Figure 13. Agricultural road planning
Proc. of SPIE Vol. 8526 852608-9
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 11/26/2012 Terms of Use: http://spiedl.org/terms
Figure 14. Orthophoto at Gunung Bongsu
1
Figure 15. Replanting planning consisting of terraces and road.
Proc. of SPIE Vol. 8526 852608-10
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 11/26/2012 Terms of Use: http://spiedl.org/terms
3.2 Optical Analysis
In order to generate a digital classified land cover map of the study area, a standard remote sensing classification
algorithm of maximum likelihood was used. Maximum likelihood is adopted due to its simplicity and ability to provide
accurate classification output [10-11].
LULC Ijok Plantation
693,0.0
69360.0
(a)
69390.0
694200
(b)
Figure 16. (a) Raw Airphoto Imagery of Ijok Plantation; (b) Classified Airphoto Imagery into LULC Map
Figure 16 (a) shows the raw airphoto imagery in RGB true colour which indicate green area as oil palm trees, light green
as nursery, grayish area as cleared oil palm area, and yellowish lines areas as terraced area. The high resolution air photo
images provided the capability for thematic mapping of land use and land cover (LULC) in oil palm plantations. In this
study, five LULC classes were mapped using the air photo using Maximum Likelihood Classifier. The LULC classes are
matured oil palm, oil palm nursery, shrubs, clearland and also water bodies. The classification outputs of the optical
airphoto give an accepted accuracy of 82.3%. This LULCs map gives plantation managers the real time LULCs of the
plantation and progress of replanting works and unexplored area that suitable for oil palm seed planting. The LULC map
also provides the boundary of cleared area that need to undergo terracing process.
Proc. of SPIE Vol. 8526 852608-11
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 11/26/2012 Terms of Use: http://spiedl.org/terms
3.3 Fusion Analysis
Terraces
Matured oil palm
Clearland
Clearland
Nursery
Figure 17. 3D model for fused data (Airphoto and LiDAR DEM)
The raw airphoto can be fused with DEM data generated from point cloud data of LiDAR. Figure 17 shows the output of
3D model generated using ENVI 4.7. Features such as matured oil palm trees, nursery, agriculture roads, cleared oil
palm area and terraced hilly areas can be seen clearly using this 3D model.
Proc. of SPIE Vol. 8526 852608-12
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 11/26/2012 Terms of Use: http://spiedl.org/terms
Terraces
Matured oil palm
Clearland
,-
Nursery
Clearland
Figure 18. LULC map fused with DEM information of study area.
Figure 18 shows the LULC map fused with DEM of the study area. The fused data provide information about terrain of
LULC located. The fused data shows that the matured oil palms are located at hilly area in the middle of the map. The
nursery is planted at low area near to the cleared land and below the matured oil palm hills. There are a mix of shrubs in
the plantations that represented by yellowish colour. Based on this information, plantation managers can have a more
complete information on their plantations and these digital information can further be manipulated for advanced GIS
analysis.
4. CONCLUSIONS
The initial remote sensing methodology for this work by principle is suitable for the planning of the replanting program.
The remote sensing-based methods are practical, straightforward and cost-effective, to be effectively utilized in the
plantation operation as well as for the design of agricultural road and terrace construction. The data could be then be
transferred into handheld GPS such as Trimble GPS receivers used for updating the existing road data layer of the Felda
plantations. These technologies were welcomed by plantation management, as remote sensing technology established a
dependable basis on which to make decisions. Plantation companies involved in the case studies considered that the
application of remote sensing would lead to greater efficiencies and therefore greater profitability. However, further
education and industry exposure is necessary to bring greater awareness of this technology to the palm oil plantation
industry. In FELDA, there still a need to create awareness among the plantation staff on how to use and manipulate the
remote sensing techniques for operation work. The approach should expand and emphasis the provision of data for the
management of reconstructing and redesigning terraces and agricultural road. High spatial resolution airphoto assisted in
providing LULC map with an accuracy of 82.3% which is adequate for managers and decision makers. The fusion of
DTM from LiDAR and the LULC map give a 3D view of current condition of the oil palm plantations. Future work
includes the automation of tree counting and disease analysis of oil palm plantations at larger scale and establishment of
Proc. of SPIE Vol. 8526 852608-13
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 11/26/2012 Terms of Use: http://spiedl.org/terms
comprehensive digital database of all plantations. With the digital database created, implementation of GPS-guided
machineries for road and terrace construction is also possible. Another interesting area of research and operation is the
combined use of hyperspectral and LiDAR systems in order to generate a more accurate and holistic oil palm plantation
management strategies.
ACKNOWLEDGEMENT
The authors would like to thank Director General of FELDA, Tuan Haji Faizoull Ahmad, Deputy Director General of
FELDA, Tuan Haji Mohd Nor Kailany, CEO of Felda Technoplant Sdn Bhd Mr Yusof bin Mohd Arshad and Chief
Operatiing Officer Of Felda Technoplant Sdn Bhd Tuan Haji Musbahudin Kasim for funding and facilities during the
study and data acquisition process.
REFERENCES
[1] Manaf, N., [The Impacts of Topography On Oil Palm Plantation]. Master Thesis, Universiti Teknologi Mara (UiTM)
(2010).
[2] The World Bank, [WDR 1994: Infrastructure for Development], World Bank and Oxford University Press, 978-0-19520992-1, doi: 10.1596/978-0-19-520992-1, (1994).
[3] Zainuddin, N.S., [Effectiveness Of Sediment Basin And Silt Traps At Oil Palms Plantations], Master Thesis,
Universiti Teknologi Malaysia (UTM), (2007).
[4] Shafri, H.Z.M, Anuar, M.I, and Saripan, M.I., “Modified vegetation indices for Ganoderma disease detection in oil
palm from field spectroradiometer data,” Journal of Applied Remote Sensing, vol. 3(1), pp. 033556-033556-23, (2009).
[5] Shafri, H.Z.M, Anuar, M.I., Seman, I.A., and Noor, N.M., “Spectral discrimination of healthy and Ganodermainfected oil palms from hyperspectral data,” International Journal Of Remote Sensing, vol 32(22), pp. 7111-7129,
(2011a).
[6] Shafri, H.Z.M., Nasrulhapiza, H., and Mohamad, I.A., “Detection of stressed oil palms from an airborne sensor using
optimized spectral indices,” International Journal of Remote Sensing,, vol.33(14), pp. 4293-4311 (2011b).
[7] Clode, S., F. Rottensteiner, P. Kootsookos, and E. Zelniker, "Detection and vectorization of roads from LiDAR data,"
Photogrammetric Engineering & Remote Sensing, vol. 73(5), pp. 517-535, (2007).
[8] Bales, J.D., Wagner, C.R., Tighe, K.C., and Terziotti, Silvia, “LiDAR-derived flood-inundation maps for real-time
flood-mapping applications, Tar River basin, North Carolina,” U.S. Geological Survey Scientific Investigations Report
2007, vol. 5032, pp.1-42 (2007).
[9] Jupp, D.L.B., Culvenor, D., Lovell, J.L., and Newnham, G., “Evaluation and validation of canopy laser radar
(LiDAR) systems for native and plantation forest inventory,” CSIRO Marine And Atmospheric Research Papers.
Number 020. Final Report to The Forest and Wood Products Research & Development Corporation, (2005).
[10] Jensen, J., [Introductory Digital Imageprocessing: A Remote Sensing Perspective], 3rd ed., Upper Saddle River, NJ,
Prentice Hall, (2005).
[11] Manandhar, R., Odeh, I.O.A., and Ancev, T., “Improving the Accuracy of Land Use and Land Cover Classification
of Landsat Data Using Post-Classification Enhancement,” Remote Sensing, vol.1, pp.330-344, (2009).
Proc. of SPIE Vol. 8526 852608-14
publicationhttp://proceedings.spiedigitallibrary.org/
stats
DownloadedViewFrom:
on 11/26/2012 Terms of Use: http://spiedl.org/terms