Proceedings of IAPR Workshop on MACHINE VISION APPLICATIONS

(IAPR MVA '94)

DECEMBER 13-15, 1994, KAWASAKI, JAPAN

Table of Contents

Front Messages in the Proceedings

Session 1: Architecture

3D Cone-beam Tomography for Non-destructive Testing(Invited) 1
Per-Erik Danielson, Linkoping Univ. , Sweden
Real-time Geometric Distortion Correction and Image Processing on the 1D SIMD architecture IVIP 9
Anders Aström** and Folke Isaksson**, *Linköping Univ. , **Saab Missiles AB, Sweden
A Compact 30GIPS Real-Time Vision System RVS-2 13
ShinIchiro Okazaki, Yoshihiro Fujita and Nobuyuki Yamashita, NEC Corp. , Japan

Session 2: 3D Object Recognition

Object Recognition Using Conic-Based Invariants from Multiple Views 17
Osamu Takae, Yoshinori Kuno, Jun Miura and Yoshiaki Shirai, Osaka Univ. , Japan
Rapid Recognition of Elementary Surface Shapes in Cluttered Range Images Using Tripod Operators 21
Frank Pipitone, Naval Research Laboratory, USA
Region Correspondence for Color Scene Images Taken from Different Viewpoints 26
Caihua Wang and Keiichi Abe, Shizuoka Univ. , Japan
Connectivity in High Dimensional Images, 30
Pieter P. Jonker and Albert M. Vossepoel, Delft Univ. of Technology, The Netherlands
A Multi-scale Approach for Crest Line Extraction in 3D Medical Images 34
Olivier Monga*, Richard Lengagne*, Rachid Deriche*, Song De Ma**, *INRIA **Academia Sinica, *France, ** China

Session 3: Poster Session 1

Represent and Acquire Knowledge for the Development of Autonomous Vision System 38
Xiaobu Yuan, Memorial Univ. of Newfoundland, Canada
Range Image Integration Based on 2D Shape Matching 44
Hitoshi Wakizako*, Sei-ichiro Kamata** and Eiji Kawaguchi**, *YASKAWA Electric Corp. , **Kyushu Institute Technology, Japan
Automated Map-Drawing Digitizing Technology Based on Scanner Input 48
S. Ablameyko, V. Bereishik and A. Kryuchkov, Belarusian Academy ofSciences, Belarus
A New Retina-Like Visual Sensor Performing the Polar Transform 52
Cheon Woo Shin and Seiji Inokuchi, Osaka Univ. , Japan
Fuzzy Relational Model with 3-D Structural Information of Objects 57
Gang Dong, Tomohiro Yamaguchi and Masahiko Yachida, Osaka Univ., Japan
A Fuzzy Reasoning Rule-Based System for Lace Pattern Detection 61
Nasser Sherkat, Chi-Hsien V. Shih and Peter Thomas, The Nottingham Trent Univ. , UK
Generating Symbolic Descriptions of Two-dimensional Blocks World 65
A. Sowmya and K. W. E. Lee, Univ. of New South Wales, Australia
Machine Vision Algorithms on a Color Blindness Plate 71
Yung-Sheng Chen and Yu-Chang Hsu, Yuan-Ze Institute of Technology, Taiwan
A High-Speed Character Contour-Fill Method Using an Edge-Flag List 75
Kazuki Nakashima*, Masashi Koga*, Katsumi Marukawa*, Yoshihiro Shima* and Yasuaki Nakano**, *Hitachi, Ltd. , **Shinshu Univ. , Japan
Video Retrieval Method Using a Sequence of Representative Images in a Scene 79
Akio Nagasaka, Takafumi Miyatake and Hirotada Ueda, Hitachi, Ltd. , Japan
Image Analysis for Automated Pavement Cracking Evaluation 83
Hideki Shimamura, Kazuhiro Oonuma, Yoshizumi Yasuda, Univ. of Chiba, Japan
From Step Edge to Line Edge: Combining Geometric and Photometric Information 87
Alain Filbois and Didier Gemmerlé, CRIN/CNRS - INRIA Lorraine, France
On a Segmentation Algorithm of Lines of Hough Transform using Discriminant Analysis - on the Enforcement of the Platform of Hough Transform - 91
Kunihito Kato, Kazuhito Murakami and Hiroyasu Koshimizu, Chukyo Univ., Japan
Skew Detection Using Directional Profile Analysis 95
Stephen W. K. Lam and Victor C. Zandy, State Univ. of New York at Buffalo, USA
A Modified Simulation Environment for Reconfigurable Multicomputer Systems in Digital Image Processing Applications 99
Francisco J. Quiles Flor and Antonio Garrido del Solo, Campus Univ. , Spain
Automatic Switching System of Rotary DIP Switch in Inspection of Digital Relay Board 103
Tetsuya Matsumoto*, Yoichi Kitagawa* , Yukako Takizawa* and Takumi Minemoto** , *Hyogo@Prefectural Institute of Industrial Research, **Kobe Univ. , Japan
Robust Line Drawing Understanding Incorporating Efficient Closed Symbols Extraction 107
Shin'ichi Satoh*, Hiroshi Mo** and Masao Sakauchi**, *National Center for Science Information Systems, ** Univ. of Tokyo, Japan
A Pixel Voting Method to Recover 3D Object Shape from 2D Images 111
Kenpo Tsuchiya*, Shuji Hashimoto* and Toshiaki Matsushima**, * Waseda Univ. , ** Toho Univ. , Japan
Synchronization between Image, Sound and Scenario Document using DP Matching 115
Yoshitomo Yaginuma, Makoto Kageyama and Masao Sakauchi, Univ. of Tokyo, Japan
A B-Spline Based Algorithm for Shape From Shading 119
Gu Junxiao* and Visakan Kadirkamanathan**, *Ministry of Astronautics Industry of China, ** Univ. of Sheffield, *China, ** UK
Curvilinear Network Extraction from Remotely Sensed Images 123
Mark R. Dobie, Paul H. Lewis and Mark S. Nixon, Univ. of Southampton, UK
Image Segmentation Using both Edge and Region Information 127
Takahiro Sugiyama and Keiichi Abe, Shizuoka Univ. , Japan
Visual Control of a Robot Head 131
Han Wang, Stan Z. Li and E. K. Teoh, Nanyang Technological Univ. , Singapore
Stroke Extraction as a Basis for Structural Analysis of Document Images by Modified MCR Expression 135
Supoj Chinveeraphan, AbdelMalek B. C. Zidouri and Makoto Sato, Tokyo Institute of Technology, Japan
3D Object Model Fitting to Still Images Using Linear Combination Method of 2D Aspect Images 139
Hiroyasu Sakamoto* and Masahide Kawakami**, *Kyushu Univ. of Design, **Namko Co. , Japan
A System for Recognizing Forms and Characters on Cylindrical Objects 143
Tapio Seppänen*, Pasi Hyttinen*, Matti Pietikäinen* , Hannu Hahtala** and Jukka Rajala**, * Univ. of Oulu, **Sypal Inc. , Finland
Obtaining 3-D shape from Silhouette Informations Interpolated by Photometric Stereo 147
Changsuk Cho, Haruyuki Minamitani, Keio Univ. , Japan
The Parallel Detection of Interesting Points in Distance Transform for Image Matching 151
J. You*'**, E. Pissaloux*, T. Dahlin** and W. P. Zhu**, *Univ. Paris, France, **The Univ. of South Australia, **Australia
A Segmentation's Technique for Volumetric Calculation of Multiple Scleroses in Nuclear Magnetic Resonance Imaging 155
S. Reboul*, A. Taleb-Ahmed*, S. Telliez*, D. Caparros-Lefebvre**, J. P. Pruvo** and J. P. Dubus***, *Lab. instrumentation du signal, de l'image et des réseaux de Calais, **Lab. de neuroradiologie de Lille, ***Lab. de mesure automatique de Lille, ***Lab. de neuroradiologie de Lille, France
Comparing Some Tools Useing Frequency Domain for the Estimation of 1-d (and 2-d) Disparity 159
Yann Yvinec and Bernard Thiesse, Laboratoire Informatique et Mathematiques Appliquées, France
Uncalibrated Stereo Vision with Pointing for a Man-Machine Interface 163
Roberto Cipolla, Paul A. Hadfield and Nicholas J. Hollinghurst, Univ. of Cambridge, UK
Evaluating the TERASYS System for Image Processing Tasks 167
Patrick D. Krolak and Patrick G. Mullins, Univ. of Massachusetts Lowell, USA

Session 4: Motion & Navigation

Efficient Algorithms for Octree Motion 172
Andrew Smith, Yoshifumi Kitamura and Fumio Kishino, ATR, Japan
Movement Tracer System Using Non-Parallel Multiple Line Detectors and High Order Correlation Analysis 178
Masashi Sakamoto, Keisuke Kameyama, Katsuyuki Kuwano, Takuso Sato, Tokyo Institute of Technology, Japan
Disparity Selection in Binocular Pursuit 182
Atsuto Maki, Tomas Uhlin and Jan-Olof Eklundh, Royal Institute of Technology, *Sweden, **Japan
An Integration Algorithm for Stereo, Motion and Color in Real-Time Applications 186
Minoru Etoh and Hiroshi Arakawa, Matsushita Electric Industrial Co. ,Ltd. , Japan
Thorough ZDF-Based Localization for Binocular Tracking 190
N. Kita*, S. Rougeaux**, Y. Kuniyoshi* and S. Sakane*, *ETL, **Institut d'Informatique d'Entreprise, * Japan **France
Virtual Gun, A Vision Based Human Computer Interface Using the Human Hand 196
James J. Kuch and Thomas S. Huang, Univ. of Illinois, USA

Session 5: Image Processing & Algorithm

Analysis and Representation of Computer Vision Systems by the Object-Process Methodology (Invited) 200
Dov Dori, TECHNION, Israel
An Efficient Agglomerative Clustering Algorithm for Region Growing 210
Takio Kurita, ETL, Japan
Adaptive Energy Function for Active Net 214
Kazuyoshi Yoshino, Mio Maki, Toshio Kawashima and Yoshinao Aoki, Hokkaido Univ. , Japan
A High Level Merging Tool in Image Segmentation Applications 218
P. Charnier, E. Ahronovitz and C. Fiorio, Univ. Montpellier, France

Session 6: Map

Segmentation of Scanned Maps in Uniform Color Spaces 222
B. Lauterbach* and W. Anheier**, ** Univ. of Bremen, Germany
Recognition of Cartographic Symbols 226
Sushil Bhattacharjee and Gladys Monagan, Swiss Federal Institute of Technology (ETH), Switzerland
On the Extraction of Various Regions in Vector Maps 230
Guofang Jiao*, Eihachiro Nakamae*, Katsumi Tadamura* and Hiroyuki Inuyama**, *Hiroshima Prefectural Univ. , **Sanei Giken Co,,Ltd. , Japan
An Interactive Map Drawing Recognition System with Learning Ability 235
Wei Lu, Masao Sakauchi, Univ. of Tokyo, Japan
Automatic Recognition of Linear features, Symbols and Textured Areas within Maps 239
David G. Elliman and Madhumita Sen-Gupta, Univ. of Nottingham, UK

Session 7: Poster Session 2

Understanding Electronic Circuit Diagrams and Linking with ORCAD 243
Shi Zesheng, Jiang Cunhong, Li Si fang and Yang Jing, Univ. of Science and Technology of China, China
Determination of End Points and Length of a Straight Line using the Hough Transform 247
M. Atiquzzaman* and M. W. Akhtar**, *La Trobe Univ. , **King Fahd Univ. of Petroleum & Minerals, Australia
Recognition of Numeric Strings with Notation Rules Using String Checking 251
Katsumi Marukawa, Kazutoshi Takakura, Taro Hayashi, Masashi Koga and Yoshihiro Shima, Hitachi, Ltd. , Japan
From Computer Vision to Document Recognition or Using Labeling Technique for Map Interpretation 255
S. Ablameyko, O. Frantskevich, Belarusian Academy of Sciences, Belarus
Using an Octree-Based RAG in Hyper-Irregular Pyramid Segmentation of Texture Volume 259
Horace H. S. Ip and Stephen W. C. Lam, City Polytechnic of Hong Kong, Hong Kong
A Fast Corner Detection Algorithm Based on Area Deviation 263
Zhou Hao and Shao Lejun, Nanyang Technological Univ. , Singapore
An Intelligent Document Understanding & Reproduction System 267
M. Sharpe, N. Ahmed and G. Sutcliffe, James Cook Univ. of North Queensland, Australia
Methods to Estimate Areas and Perimeters of Blob-like Objects: a Comparison 272
Luren Yang, Fritz Albregtsen, Tor Lønnestad and Per Grøttum, Univ. of Oslo, Norway
A Statistically Unbiased Method for Computing the Rotation Angles from Image Sequences 277
Han Wang, Stan Z. Li and E. K. Teoh, Nanyang Technological Univ. , Singapore
Reconstruction of 3-D Terrain Data from Contour Map 281
Jin-Seon Lee and Seung-Jong Chung, Chonbuk National Univ. , Korea
Direct Reconstruction of Planar Surfaces by Stereo Vision 285
Yasushi Kanazawa* and Kenichi Kanatani**, *Gunma College of Technology, **Gunma Univ. , Japan
Directional Voting with Geometric Hashing for Three Dimensional Object Recognition 289
Shun'ichi Kaneko, Masahiro Shibata and Tsunenori Honda, Tokyo Univ. of Agriculture and Technology, Japan
Use of Three-Dimensional Templates for Multiple Skeletons 293
Masato Masuya and Junta Doi, Univ. of Tokyo, Japan
A study of Joint Set Modeling Based on Edge-Base Stereo Matching 297
Takeshi Doihara*, Tatsuo Ikeda*, Kazuo Oda*, Masahiro Idogaki**, Yasuo Uchita**, Kazuhiko Onuma*** and Yoshizumi Yasuda***, *Asia Air Survey Co. ,Ltd. , ** The Kansai Electric Power Co. ,Inc. , ***Ckiba Univ. Japan
Vector-based Editing Method of Drawings for Facility Maintenance 301
Hideaki Maehara, Jun'ichi Shibayama, Jun Sawamoto, Mitsubishi Electric Corp. , Japan
Analyses of Graphical Features of Textiles by Human Impressions 307
Utsuno Naoki, Toru Inohara, Yuudai Morohara, Kunio Kondo, Hisashi Sato and Shizuo Shimada, Saitama Univ. Japan
The Recognition of Printed Korean Characters by ART-Based Neural Network Hierarchy 311
Yong Tae Woo*, Ji Hyun Ki**, Bong Goo Lee* and Narn Il Lee**, *Changwon National Univ. , **Andong National Univ. , Korea
3D Modelling of In-Door Scenes Using Laser Range Sensing 315
Vitor Sequeira*, João G. M. Gonçalves* and M. Isabel Ribeiro**, *European Commission Joint Research Centre, **Instituto Superior Técnica, *Italy, ** Portugal
Generation of a Position Tolerant Representation of Edges 319
B. Mertsching*, J. Schnusenberg*, S. Neusser** and T. Schwederski**, * Univ. of Hamburg, **Institute for Microelectronics Stuttgart, Germany
3D-Environment Measurement System for Large-Scale Structure Manufacturing 325
Toshiyuki Aoki* and Kohji Kamejima**, *Hitachi,Ltd. , **Osaka Institute of Technology, Japan
Avoiding Redundant Processing in Gradient Based Edge Detection 330
Martin J. J. Scott and Richard W. Prager, Cambridge Univ. , UK
A Method of Understanding Conceptual Diagrams 334
Noriyoshi Yoneda, Koichi Kise, Shinobu Takamatsu and Kunio Fukunaga, Univ. of Osaka Prefecture, Japan
Employing Symmetric Subsets for Identifying Asymmetry of Human Skulls 338
Teruhito Takeda, Seiji Ishikawa and Kiyoshi Kato, Kyushu Institute of Technology, Japan
The Mask Tuning Scheme for Texture Feature Extraction and Its Parallel Implementation 342
J. You*,**, H. A. Cohen*** and E. Pissaloux*, * Univ. Paris, **The Univ. of South Australia, ***La Trobe Univ. , *France, **,***Australia
Repeated Sampling to Improve Classifier Accuracy 346
Jiangying Zhou and Daniel P. Lopresti, Panasonic Technologies, Inc. , USA
Obtaining Environment Model Using Behavior-Based Robot System 352
Koji Kato, Hiroshi Ishiguro and Saburo Tsuji, Osaka Univ. , Japan
Evaluation of Effectivity of Omnidirectional Image Sensor COPIS in a Real Environment 356
Ryota Hiura, Yasushi Yagi and Masahiko Yachida, Osaka Univ. , Japan
An Algorithm for Understanding of Color Vision 360
T. J. Fang, Y. Z. Chen, L. M. Tao, G. Z. Yao and Y. J. Wang, Academia Sinica, China
Automated Building Height Extraction and Building Detection from High Resolution Aerial and Space Imagery 364
Taejung Kim and Jan-Peter Muller, Univ. College London, UK
Image Separation between Natural and Artificial Objects Using Fractal Dimension 368
Shuichi Fukuda, Tokyo Metropolitan Institute of Technology, Japan
Off-line Handwriting Recognition by Statistical Correlation 371
G. Saon, A. Belaçd and Y. Gong, CRIN/CNRS - INRIA Lorraine, France

Session 8: Feature & Texture

Texture Features from Gray level Gap Length Matrix 375
Wang Xinli*, Fritz Albregtsen* and Bent Foyn**, * Univ. of Oslo, ** The Norwegian Pulp and Paper Research Institute, Norway
A Coarse to Fine Search Technique to Detect Circles in Images 379
M. Atiquzzaman, La Trobe Univ. , Australia
Mesh Pattern Recognition Using Correlation Matching Method, 383
Chin-Chuan Han and Kuo-Chin Fan, National Central Univ. , Taiwan
An Automatic Tree Ring Measurement System 387
Wang Xinli*,**, Bent Foyn* and Fritz Albregtsen**, *The Norwegian Pulp and Paper Research Institute, **Univ. of Oslo, Norway
Edge Extraction Method Based on Separability of Image Features 391
Kazuhiro Fukui, Toshiba Corp. , Japan
Hierachic Texture Classification Using Morphological Gradients and Genetic Algorithms 397
Po-Long Tian, Jia-Hong Lee and Yuang-Cheh Hsueh, National Chiao Tung Univ. , Taiwan

Session 9: 3D Shape Analysis

Solving Inverse Problems in Computer Vision by Scale Space Reconstruction 401
A. G. Jones and C. J. Taylor, Univ, of Manchester, UK
Adaptive Contour Model Using Texture Feature Vectors 405
Roman Durikovic, Kazufumi Kaneda, Hideo Yamashita, Hiroshima Univ,, Japan
3D Shape Reconstruction of Unfolded Book Surface from a Scanner Image -Shape from Shading with Interreflections under Proximal Light Source 409
Hiroyuki Ukida, Toshikazu Wada and Takashi Matsuyarna, Okayama Univ. , Japan
3D Information Acquisition from Spatiotemporal Image Created by a Hyperbolic Slit 413
Chunxiao Li*, Heitou Zen** and Masao Sakauchi*, *Univ. of Tokyo, **Tokyo Univ. of Mercantile Marine, Japan
An Autonomous Three-Dimensional Vision Sensor with Ears 417
Shigeru Ando, Univ. of Tokyo, Japan

Session 10: Document

Reconstruction of 3D Objects from Three Orthographic Projections Using a Decision Chaining Method 423
R. E. Marston and M. H. Kuo, Univ. of Nottingham, UK
A Procedure for Segmenting Touching Numbers in Cadastral Maps 427
Gladys Monagan, Swiss Federal Institute of Technology (ETH), Switzerland
A Local-to-Global Approach to Complex Document Layout Analysis 431
Stephen W. K. Lam, State Univ. of New York at Buffalo, USA
An Efficient OCR Error Correction Method for Japanese Text Recognition 435
Toru Hisamitsu, Katsumi Marukawa, Yoshihiro Shima, Hiromichi Fujisawa and Yoshihiko Nitta, Hitachi, Ltd. , Japan
An Algorithm to Restore a Curved Object from Three Orthographic Views by Using Probabilistic Relaxation Matching Method 439
Tomoyuki Iwama*, Takahiko Horiuchi*, Kazuo Toraichi*, Hiromitsu Yamada** and Kazuhiko Yamamoto**, * Univ. of Tsuhuba, **ETL, Japan

Session 11: Poster Session 3

The Recognition of the Composition of Paintings by Chinese Flower and Bird Painters 443
Da-Yu Liu and Yoshinao Aoki, Hokkaido Univ. , Japan
A High-Speed Document Image Classifier 447
Shao Lejun, Nanyang Technological Univ. , Singapore
Analysis System of Mixed Documents Consisting of Handwritten Korean/Alphanumeric Texts and Graphic Images 451
Chang Bum Lee*, Dong-Gyu Si***, In Kwon Kim**, Young Kug Ha*** and RaeHong Park**, *Electronics and Telecommunications Research Institute, **Sogang Univ. , Korea
3-D Shape Reconstruction from Endoscope Image Sequences by the Factorization Method 455
Koichiro Deguchi, Tsuyoshi Sasano, Himiko Arai and Yutaka Yoshikawa, Univ. of Tokyo, Japan
Image Segmentation Based on Fuzzy Clustering Algorithm 460
Chaur-Heh Hsieh, Chung-Ming Kuo, Chung-Woei Chao and Po-Chiang Lu, Chung Cheng, Institute of Technology, Taiwan
Derivative-Based Optical Flow Estimation: Controlled Comparison of First- and Second- Order Methods 464
Jonathan W. Brandt, JAIST, Japan
Image Complexity Analysis for Self-Tuning Pattern Regeneration in Open Environment Knowledge Projection 470
Kohji Kamejima*, Masakazu Ejiri** and Yuriko C. Watanabe**, *Osaka Institute of Technology, **Hitachi, Ltd. , Japan
Cooperation of 3D Segments and 3D Facets information for Object Reconstruction 474
Didier Gemmerlé and Alain Filbois, CRIN/CNRS - INRIA Lorraine, France
A Thresholding Approach for Segmenting Deteriorated SEM Images in an Automated X-ray Mask Visual Inspection System 478
Minoru Ito, Kogakuin Univ. , Japan
A Processor Architecture for Discrete Wavelet Transform Processing of Gray Level Image. 482
P. E. Indrajit Dimyati and Tati L. R. Mengko, Institut Teknologi Bandung (ITB), Indonesia
A Novel Approach for Detection of Edges in Range Images Using Splines 486
Satish Kaveti, Earn Khwang Teoh and Han Wang, Nanyang Technological Univ. , Singapore
Development of Surface Inspection Machine for Organic Photo Conductor(OPC) 490
Osamu Nakayama, Shinji Kobayashi, Katsuyuki Omura, Takahiro Asai, Mitsuhiro Tomoda and Teruki Kamada, Ricoh Co. , Ltd. , Japan
Machine Vision Based Inspection of Textile Fabrics 494
Chuanjun Wang and Chih-Ho Yu, Tsinghua Univ. , China
A Kernel System for 3D GIS Applications 498
Ying Ming Lin*, Yutaka Ohsawa**, Masao Sakauchi*, * Univ. of Tokyo, **Saitama Univ. , Japan
A Robust Approach for Camera Break Detection in Color Video Sequence 502
John Chung-Mong Lee and Dixon Man-Ching Ip, The Hong Kong Univ. of Science & Technology, Hong Kong
Robust Curvature Vectors Calculation from Range Data Using ISL Method506 506
Takashi Yoshimi and Fumiaki Tomita, ETL, Japan
An Adaptive Welding System for Corner Joints Which Applies Joint Recognition by Deformable Template Matching 510
Ryosuke Mitaka, Matsushita Electric Works, Ltd. Japan
The Evaluation of Textile Design's Visual Feature by Using a Genetic Algorithm 514
Ken'ichi Ohta*, Kouichi Nishida*, Fujio Miyawaki* and Katsuhiko Sakaue**, *Himeji Institute of Technology, **ETL, Japan
Stable Position Measurement of a Moving Airplane by Adaptive Camera Control 518
Yuichi Togashi and Hideaki Yamaguchi, Toshiba Corp. , Japan
The Effects of Document Image Defects on Line Drawing Analysis Algorithms 522
Yuh-Lin Chang and Daniel P. Lopresti, Panasonic Technologies, Inc. , USA
Transferring Human Sensibilities to Machines -Sensitivity Analysis of Layered Neural Networks and Its Application to Pearl Color Evaluation 528
Noriko Nagata, Mitsuhito Kamei and Teruo Usami, Mitsubishi Electric Corp. , Japan
A Study on Underwater Cable Automatic Recognition Using Hough Transformation 532
Chao Chen and Masayuki Nakajima, Tokyo Institute of Technology, Japan
A Human Detector Based on Flexible Pattern Matching of Silhouette Projection 536
Hajime Ohata*, Nobuyoshi Enomoto*, Akio Okazaki*, Hiroaki Kawasumi**, Shigeo Sudo**, Yoshiaki Yamada**, * Toshiba Corp. , ** Tokyo Electric Power Company, Japan
Shape Identification Using Probabilistic Models of Attentional Sequences 540
H. Isil Bozma and Çagatay Soyer, Bogaziçi Univ. , Turkey
A Visual Navigation System for Autonomous Underwater Vehicle 544
Wang Jian-gang, Hao Ying-ming, Cao Hong-kai, Yu Yong and Xu Xin-ping, Chinese Academy of Sciences, China
A Fast Boundary Based Thinning Algorithm 548
Y. Zhu, L. D. Seneviratne and S. W. E. Earles, King's College London, UK
Document Degradation Models: Parameter Estimation and Model Validation 552
Tapas Kanungo*, Robert M. Haralick*, Henry S. Baird**, Werner Stuetzle* and David Madigan*, * Univ. of Washington, **AT&T Bell Laboratories, USA

Session 12: Applications

Detection of Defects in Colour Texture Surfaces (Invited) 558
J. Kittler, R. Marik, M. Mirmehdi, M, Petrou and J. Song, Univ. of Surrey, UK
A Digital ULSI Inspection Method Using Parallel Scanning Confocal Microscope 568
Yoshiharu Ichikawa**** and Jun-Ichiro Toriwaki**, * Texas Instrument Japan Ltd. , **Nagoya Univ., Japan
Three-dimensional Measurement Approach for Automatic Seal Impression Identification 571
Ryoji Haruki* , Takahiko Horiuchi* , Kazuo Toraichi*, Kazuhiko Yamamoto**, Hiromitsu Yamada** and Anis Fu*, * Univ. of Tsukuba, **ETL, Japan
Automated Inspection System for Electronic Devices Mounted on Printed Circuit Boards Using Light-Section Method 575
Toshifumi Honda, Hisae Yamamura, Syouji Ueda, Minoru Kawabata and Takanori Ninomiya, Hitachi,Ltd. , Japan
Image Metamorphosis Using Optical Flow Techniques 579
Minas E. Spetsakis, York Univ., Canada