Newsletter 05

Prof. Kazuhiko Yamamoto, general chair of the 2nd ICDAR
(This email address is being protected from spambots. You need JavaScript enabled to view it. ), kindly sent me an electronic version of the
ICDAR'93 advance program and asked me to distribute it. That's the
reason for this new mailing just a day after the last one. For the
benefit of all those who haven't received it through another
channel... Please feel free to circulate it to all those who might be

Karl Tombre.
[From the conference Secretariat V.3]

This article is an electronic version of the Advance Program of ICDAR'93
(Second International Conference on Document Analysis and Recognition),
which will be held on October 20-22, 1993 at Tsukuba, Japan.

If you want a printed copy of the program as well as those of the Registration
Form and the Hotel Reservation Form, please contact:
ICDAR'93 Conference Secretariat
c/o Prof. K. Yamamoto, Head of Image Understanding Section
Electrotechnical Laboratory
1-1-4 Umezono, Tsukuba
Ibaraki 305, JAPAN
E-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
Fax: +81-298-58-5949

This electronic version does not contain the Registration Form and the
Hotel Reservation Form. The Secretariat will send them on demand via ordinary
mail. If you have a postscript printer, you are recommended to obtain the
postscript file of the images via e-mail. Please send a request mail to the
E-mail address shown above.

Contents of this article:
[Organizing Committee]
[Executive Committee]
[Program Committee]
[Extended Executive Committee]
[Social Events]
[Commemorative CD-ROM]
[Location of the Conference]
[Technical Sessions]
[Poster Session]
[Ground Transportations]

Advance Program
Second International Conference on Document
Analysis and Recognition

Tsukuba Dai-ichi Hotel
Tsukuba Science City, Japan
October 20 - 22, 1993

Sponsored by IAPR TC-11, TC-10 and IGS
In cooperation with IEEE-CS

[Organizing Committee]
General Chair:
K. Yamamoto, Electrotechnical Laboratory
General Co-chairs:
C. Y. Suen, Concordia University, Canada
M. Sakauchi, The University of Tokyo
Program Chairs:
T. Pavlidis, State University of New York at Stony Brook, U.S.A.
S. Mori, The University of Aizu
Y. Nakano, Shinshu University
Executive Chairs:
H. Fujisawa, Hitachi, Ltd.
K. Sakai, Toshiba Corporation
J. Tsukumo, NEC Corporation
Public Relations:
H. Asada, Toshiba Corporation
K. Ejiri, Ricoh Co. Ltd.
Local Arrangement:
K. Toraichi, University of Tsukuba
T. Nireki, Japan Electronic Industry Development Association
Advisory Committee:
T. Iijima, Japan Advanced Institute of Science and Technology, Hokuriku
M. Nagao, Kyoto University

[Executive Committee]
S. Hayashi, SANYO Electric Co., Ltd.
T. Kato, Electrotechnical Laboratory
S. Katoh, IBM Japan, Ltd.
T. Kawatani, Hewlett-Packard Laboratories Japan
Y. Kimura, CANON INC.
T. Morita, SHARP Corporation
H. Nambu, Mitsubishi Electric Corporation
Y. Shima, Hitachi, Ltd.
T. Temma, NEC Corporation
T. Wakahara, Ministry of Posts and Telecommunications
Y. Yamashita, Oki Electric Industry Co., Ltd.
E. Yodogawa, NTT Basic Research Laboratories

[Program Committee]
K. Abe, Shizuoka University
T. Akiyama, NTT Human Interface Laboratories
N. Babaguchi, Osaka University
H. S. Baird, AT&T Bell Laboratories, U.S.A.
G. Borgefors, SDRE, Sweden
R. G. Casey, IBM Almaden Research Center, U.S.A.
T. Ejima, Kyushu Institute of Technology
R. M. Haralick, University of Washington, U.S.A.
J. S. Huang, Academia Sinica, Chinese Taipei
S. Impedovo, Universita degli Studi di Bari, Italy
K. Inaba, Geographical Survey Institute
S. Kakumoto, Hitachi, Ltd.
R. Kasturi, Pennsylvania State University, U.S.A.
J. H. Kim, KAIST, Korea
J. Kittler, University of Surrey, U.K.
S.-W. Lee, Chungbuk National University, Korea
Guy Lorette, Universite de RENNES 1, France
M. Minoh, Kyoto University
S. Naito, NTT Basic Research Laboratories
M. Nakagawa, Tokyo University of A&T
T. A. Nartker, University of Nevada, Las Vegas, U.S.A.
H. Ogawa, Tokyo Institute of Technology
W. G. Oh, SERI/KIST, Korea
Y. Ohsawa, Saitama University
S. Ozawa, Keio University
R. Plamondon, E. Polytechnique de Montreal, Canada
A. Rosenfeld, University of Maryland, U.S.A.
J. Schuermann, Daimler-Benz, Germany
J.-C. Simon, Universite P. et M. Curie, France
S. N. Srihari, SUNY at Buffalo, U.S.A.
J. W. Tai, Chinese Academy of Sciences, China
S. Tanaka, Mitsubishi Electric Corporation
Z. Tang, Tsinghua University, China
A. Thomassen, University of Nijmegen, The Netherlands
H. Tominaga, Waseda University
J. Toriwaki, Nagoya University
Y. Tsuji, NEC Corporation
S. Tsujimoto, Toshiba Corporation
H. Yamada, Electrotechnical Laboratory
S. Yamakawa, NTT Intelligent Technology Inc.
M. Yoshimura, Chubu University

[Extended Executive Committee]
T. Akiyama, NTT Human Interface Laboratories
K. Higuchi, Oki Electric Industry Co., Ltd.
H. Honda, Japan Newspaper Publishers and Editors Association
E. Ishiguro, Mitsukoshi
T. Iwahara, SANYO Electric Co., Ltd.
M. Kugai, CANON INC.
M. Kushibiki, Hitachi, Ltd.
H. Matsumura, Media Drive Laboratory, Inc.
H. Miyao, Shinshu University
N. Otsuka, University of Tsukuba
T. Saito, Electrotechnical Laboratory
G. Sato, Ricoh Company, Ltd.
Y. Tanabe, Toshiba Corporation
S. Uchiyama, Japan Electronic Industry Development Association
M. Ueda, Consultant, Information Processing
S. Yamashita, Seiko Instruments Inc.
M. Yasuda, Meisei University
F. Yoda, Mitsubishi Electric Corporation

You are cordially invited to participate in the Second International
Conference on Document Analysis and Recognition, ICDAR'93, to be held
on October 20 - 22, 1993 in Tsukuba Science City, Japan.
Automatic intelligent processing of documents and multimedia information
is at the intersection of many fields of technology, especially of computer
science, image analysis, pattern recognition, artificial intelligence,
computer linguistics, and so on.
The conference provides an international forum for research and development
people to exchange ideas and issues in these fields, and to encourage novel
system and application developments.
It includes invited talks, presentations of refereed papers and poster
presentations as well as demonstrations of presented work.
The conference is sponsored by the International Association for Pattern
Recognition (IAPR) Technical Committees 10 and 11, and International
Graphonomics Society (IGS), in cooperation with IEEE-CS. The supporting
organizations are IEICE, IPSJ, IEEJ, JEIDA, JNNS and AVIRG. Financial
supports are given by the Tsukuba EXPO'85 Memorial Foundation and also
by Japanese industrial firms.

Character segmentation and recognition
Document structure analysis
Map and drawing interpretation
Document image processing
On-line recognition
Hypermedia processing
Neural network and psychological approaches
Implementations, applications and systems
Research environments

Technical Sessions-------------------------------------------------
Over 250 technical papers were received by the program committee. They were
>from 23 countries, in almost equal numbers from Asia, Europe and North America.
There are 28 sessions in total covering about 150 oral presentations and 66
posters. Oral presentations include five invited papers, 20 long papers and
124 short papers, marked as (I), (L) and (S) in the program. They will be
given 30, 25 and 15 minutes, respectively. These presentations are in three
parallel sessions.
A keynote speech (K) and two invited talks (T) will be given in the plenary
sessions by internationally distinguished speakers from the fields of
multimedia, pattern recognition and document analysis.

The organizing committee provides several SUN Sparc workstations and several
IBM PCs with basic operating systems at the conference site for the conference
The accepted paper/poster presenters may make demonstrations of their system
and/or algorithm in the time and machine constraints. Volunteers may
communicate with the organizing committee to allocate a time slot and to
negotiate the machine conditions.

Keynote Speech-----------------------------------------------------
Norman Meyrowitz, Director of System/User Software at GO Corporation will talk
on "Personal Communicators, Handwriting, Hypertext, and the New Mobility." The
talk will cover the emerging market of handheld, mobile devices, and examine
how handwriting recognition and hypertext/electronic book technologies will
be merged to create a personal information and communications device that
could become the telephone of the 21st century.
He directs the groups that develop the graphics, handwriting, application
frameworks, system applications, user interface and international components
of the PenPoint operating system. Most recently, he was in charge of the
development effort for the Japanese version of PenPoint.
Before joining GO, he was Co-Director of Brown University's Institute for
Research in Information and Scholarship (IRIS), where he was the principal
architect of Intermedia system, a networked, shared, multi-user hypermedia
system for research and education. He has served on the program committees
of a variety of conferences including Hypertext'91.

Invited Talk I-----------------------------------------------------
Dr. Toshiyuki Sakai, Dean, Faculty of Science and Technology, Ryukoku
University will give a talk titled "A History and Evolution of Document
Information Processing." As a pioneer of the fields of speech and character
recognition and image processing, he talks about the past and the future of
document information processing from his original viewpoint.
Starting from the analysis of speech sounds for the telephone systems, he
expanded his work into character recognition in 1960s. He has devoted himself
to the researches on pattern understanding of speech, images and documents.
He has been one of the most influential leaders of the field in Japan.
He served for various academic associations and the government as various
kinds of committee member. Among these are the presidents of IPSJ (1983-85)
and IAPR (1984-86). He received numerous awards including Imperial Purple
Ribbon Medal. After his retirement from Kyoto University in 1988, he is also
Professor Emeritus, Kyoto University.

Invited Talk II----------------------------------------------------
Prof. Dr. Andreas Dengel, Scientific Director of the German Research Center
for Artificial Intelligence (DFKI), Germany, will talk on "The Role of
Document Analysis and Understanding in Multi-Media Information Systems." He
observes that one obstacle for the development of general and homogeneous
information systems is the lacking possibilities for combining the information
processing capabilities of the electronic medium and traditional ways in which
information is generated, interchanged, categorized, processed or stored.
He will outline the role and the potential of the document analysis and
understanding technology to overcome this problem. He will also give an
overview of their efforts focusing on the integration of such techniques into
active interfaces for multimedia information systems.
Dr. A. Dengel is also a Professor in the Computer Science Department at the
University of Kaiserslautern. He received the Ph. D. degree from the
University of Stuttgart in 1989. In 1988, his work was honored by an award
>from the German Society for Pattern Recognition (DAGM).

[Social Events]
International Panel Discussion-------------------------------------
A panel discussion by three world-famous scholars is planned to be held as
a pre-opening event from 17:30 through 20:00 on October 19 at Nova Hall. The
Hall is adjacent to Tsukuba Dai-ichi Hotel which is the main venue of the
conference and it has the capacity of 1000 people. The admission is free
and open to the public.
The panelists are Dr. Leo Esaki, President of Tsukuba University, a Nobel
Prize laureate, Dr. Jean-Claude Simon, Professor Emeritus, Universite P. et
M. Curie, France, and Prof. Makoto Nagao, Kyoto University.
The main theme of the discussion will be "Evolution and Innovation - The
Frontiers of Science and Technology in the Age of Information Society."
Each of the panelists will make a keynote speech at first and then receive
questions and discussions from the floor. Dr. L. Esaki will first talk on
"Creativity and originality in the basic research" based on his long career
of research activities in the United States and Japan.
Dr. J.-C. Simon will talk about "Chaos and flexibility in the human vision
and machine vision." His research areas have been artificial intelligence
and pattern recognition. Recent work includes handwritten cursive word
recognition. He has been one of the most influential leaders of this field
in France.
Prof. M. Nagao will talk on "Artificial intelligence and pattern recognition"
based on his research activities in the areas of pattern recognition, image
processing, natural language processing and machine translation. He is one
of the pioneers in the field of machine translation especially.

Technical Tour-----------------------------------------------------
As Tsukuba is a high-tech, science city, a laboratory tour visiting the
following institutions will be held on Tuesday, October 19.
Electrotechnical Laboratory
University of Tsukuba
Real World Computing Partnership Tsukuba Research Center
RWCP is a new research organization to carry out the "Real World Computing
Program" which has been established by the Japanese government in 1992.
The tour bus will start at Tsukuba Dai-ichi Hotel at 13:00 and return there
before 17:00. You may join the international panel discussion. The tour is
free of charge but a reservation is required. For reservation, please indicate
it in the registration form.

Spouse Programs----------------------------------------------------
The following programs are planned to let you feel "real" Japan. They are
free of charge but a reservation is required for each. For reservation,
please indicate it in the registration form.
Thursday, October 21
Full Day Tour: 9:00 - 15:00
You will visit Kasama-Inari Shrine that is famous for its chrysanthemum
flowers that are best in October. You may enjoy pottery workshops around
the area too.
Friday, October 22
Half Day Tour: 9:00 - 12:00 noon
The tour will include a tea ceremony and flower arrangement demonstrations at
an old Japanese house in Tsukuba.

[Commemorative CD-ROM]
The ICDAR'93 commemorative version of CD-ROM databases will be available at
the site of the conference. There are four kinds of CD-ROMs that are
collections from the character sets of ETL-1 through ETL-9 holding 1.6
million character samples, and a Japanese document image database.
CD-ROM-A: ETL-1 and ETL-2; handwritten alphanumerics and printed kanji's.
CD-ROM-B: ETL-3 thru ETL-7; handwritten alphanumerics, symbols, hira-kana's
and kata-kana's. CD-ROM-C: ETL-8 and ETL-9; handwritten kanji's. CD-ROM-D:
Document page images from ten different kinds of documents, mostly Japanese,
including newspaper (A2 full-page in 600 dpi), novels, technical papers,
patent gazette, etc.
Registrants including students may obtain one of the four kinds of CD-ROMs
for free. See another section to see how to apply.

[Location of the Conference]
The location, Tsukuba Science City, is a well-known academic, high-tech area
in Japan. It is about 60 km north-east of metropolitan Tokyo. Foods and
accommodations are much less expensive there than in Tokyo.
The conference's main venue is Tsukuba Dai-ichi Hotel whose address is
Tsukuba Center Buildings, Azuma 1-chome, Tsukuba 305, Japan. Registration,
technical sessions, demonstrations and banquet will be in the hotel. The
plenary sessions, the poster session and the pre-opening event will take
place at Nova Hall which is adjacent to Dai-ichi Hotel.
The ground transportation from the Narita International Airport to the hotel
is explained in another section.

(All technical sessions 1-1-A thru 3-4-C are in Annex Subaru-A, -B and -C.)
Tuesday, October 19--------------------------------------------------------
16:00-18:00 Registration Annex Lobby
17:30-20:00 Nova Hall
Panel Discussion by Dr. L. Esaki, Prof. J.-C. Simon and Prof. M. Nagao
20:00-21:00 Registration Annex Lobby
Wednesday, October 20------------------------------------------------------
8:30-10:00 Registration Annex Lobby
10:00-11:15 Session 1-1-A Handwritten Character Recognition I
Session 1-1-B Feature Extraction
Session 1-1-C Image Enhancement and Bar Codes
11:30-13:00 Session 1-2-A Model-Directed Character Recognition
Session 1-2-B Layout Analysis I
Session 1-2-C Map Interpretation I
13:00-14:15 Lunch
14:15-16:00 Session 1-3-A HMM Based Approaches
Session 1-3-B Classification and Pre-classification I
Session 1-3-C On-Line Character Recognition
16:30-17:00 Plenary Session: Award Presentations Nova Hall
17:00-18:00 Nova Hall
Keynote Speech by Norman Meyrowitz, GO Corporation
18:00-20:00 Reception Saiun & Shiun
Thursday, October 21-------------------------------------------------------
9:00-10:45 Session 2-1-A Contextual Processing I
Session 2-1-B Neural Networks
Session 2-1-C Thinning and Thresholding
11:00-13:0 Session 2-2-A Cursive Word Recognition
Session 2-2-B Layout Analysis II
Session 2-2-C Interpretation of Drawings
13:00-14:15 Lunch
14:15-15:15 Nova Hall
Invited Talks by Prof. T. Sakai and Prof. Dr. A. Dengel
15:30-17:30 Poster session Nova Hall
18:30-20:30 Banquet Annex Subaru
Friday, October 22---------------------------------------------------------
9:00-10:45 Session 3-1-A Handwritten Character Recognition II
Session 3-1-B Robust Recognition of Printed Documents
Session 3-1-C Systems and Applications
11:00-13:00 Session 3-2-A Databases for OCR
Session 3-2-B Analysis of Forms, Equations and Tables
Session 3-2-C Hypermedia and Content Retrieval
13:00-14:15 Lunch
14:15-15:45 Session 3-3-A Segmentation of Touching Characters
Session 3-3-B Classification and Pre-classification II
Session 3-3-C Component Recognition in Drawings
16:00-17:15 Session 3-4-A Contextual Processing II
Session 3-4-B Layout Generation and Decomposition
Session 3-4-C Map Interpretation II
17:15 Conference Ends

[Technical Sessions]
Wednesday October 20-------------------------------------------------------
10:00 - 11:15
Session 1-1-A Handwritten Character Recognition I
(S) An Extended-Shadow-Code Based Approach for Off-Line Signature Verification
Robert Sabourin, Mohamed Cheriet and Ginette Genest (CANADA)
(S) Extended Relaxation Matching, Including DP Matching
T. Horiuchi, K. Toraichi, K. Yamamoto and H. Yamada (JAPAN)
(S) Design and Implementation of Optimized Nearest Neighbor Classifiers for
Handwritten Digit Recognition
(S) Handprinted Numeral Recognition with The Learning Quadratic Discriminant
Takahiko Kawatani (JAPAN)
(S) Improvements of A Lexicon Directed Algorithm for Recognition of
Unconstrained Handwritten Words
F. Kimura, M. Shridhar and Z. Chen (JAPAN)

Session 1-1-B Feature Extraction
(S) Primitives segmentation and association for a line character
Yung-Sheng Chen (CHINESE, Taipei)
(S) Structural Feature Extraction on Multiple Bases with Application to
Handwritten Character Recognition Systems
Hirobumi Nishida (JAPAN)
(S) A Multi-Stage Algorithm of Extracting Joint Points for Generating
R. Haruki, Y. Ohtaki and K. Toraichi (JAPAN)
(L) Stroke Segmentation as a Basis for Structural Matching of Chinese
F. Chang, Y,-C. Chen, H.-S. Don, W.-L. Hsu and C.-I. Kao
(CHINESE, Taipei)

Session 1-1-C Image Enhancement and Bar Codes
(S) Bar Code Localization Using Texture Analysis
Anil K. Jain and Yao Chen (USA)
(S) A Bar Code Location Algorithm Using a Two-Dimensional Approach
Christian Viard-Gaudin (FRANCE)
(S) Document Skew Detection Based on Local Region Complexity
Yasuto Ishitani (JAPAN)
(S) Image Continuation
Nina Billawala, Peter E. Hart and Mark Peairs (USA)
(S) A Character Image Enhancement Method from Characters with Various
Background Images
Hideaki Ozawa and Toru Nakagawa (JAPAN)

11:15 - 11:30 Break

11:30 - 13:00
Session 1-2-A Model-Directed Character Recognition
(I) Document Image Defect Models and Their Uses
Henry S. Baird (USA)
(S) A Model-Based Classifier in A Scheme of Recognition Filter
Yu Nakajima and Shunji Mori (JAPAN)
(S) A Markov Language Model in Handwritten Chinese Text Recognition
Hsi-Jian Lee and Che-Hui Chang Chien (CHINESE, Taipei)
(S) Making a Personal Recognition Dictionary from Characters Automatically
Generated by Using Handwriting Model
Nobuyuki Kita (JAPAN)
(S) An Elastic Matching Approach Applied to Digit Recognition
Jean-Michel Bertille (FRANCE)

Session 1-2-B Layout Analysis I
(S) Initial Learning of Document Structure
Andreas Dengel (GERMANY)
(S) A Block Segmentation Method for Document Images with Complicated Column
Yuki Hirayama (JAPAN)
(S) Document Recognition/Authentication Based on Medium-Embedded Random
D. Brzakovic and N. Vujovic (USA)
(S) Document Structures: A Survey
Yuan Y. Tang and Ching Y. Suen (CANADA)
(S) Document Architecture Language (DAL) Approach to Document Processing
Chiu L. Yu, Yuan Y. Tang and Ching Y. Suen (CANADA)
(S) An Intelligent Document Understanding System
Suzanne Liebowitz Taylor, Mark Lipshutz,
Deborah A. Dahl and Carl Weir(USA)

Session 1-2-C Map Interpretation I
(S) Color Map Image Segmentation Using Optimized Nearest Neighbor Classifiers
(S) Computer-Aided Cartographical System for Map Digitizing
S. Ablameyko, B. Beregov and A. Kryuchkov (BELARUS)
(L) Drawing Image Understanding System with Capability of Rule Learning
Shin'ichi Satoh and Masao Sakauchi (JAPAN)
(S) Evaluation Method for an Automatic Map Interpretation System for Cadastral
Rik D. T. Janssen, Robert P. W. Duin and Albert M. Vossepoel
(S) Automatic Lettering of Cadastral Maps
V. Consorti, L. P. Cordella and M. Iaccarino (ITALY)

13:00 - 14:15 Lunch

14:15 - 16:00
Session 1-3-A HMM Based Approaches
(L) Detecting and Locating Partially Specified Keywords in Scanned Images
Using Hidden Markov Models
Francine R. Chen, Lynn D. Wilcox and Dan S. Bloomberg (USA)
(S) Sophisticated Topology of Hidden Markov Models for Cursive Script
A. Kaltenmeier, T. Caesar, J. M. Gloger and E. Mandler (GERMANY)
(S) Handwritten Hangul Recognition with Multiple Stochastic Models
Hee-Seon Park and Seong-Whan Lee (KOREA)
(S) A Statistical Approach with HMMs for On-Line Cursive Hangul (Korean Script)
Bong Kee Sin and Jin Hyung Kim (KOREA)
(S) Reading Constrained Text Using Hierarchical Hidden Markov Models
Stephen W. Lam and Waikwan Hui (USA)
(S) Joint Normalization and Recognition of Degraded Document Images Using
Pseudo-2D Hidden Markov Models
Oscar E. Agazzi and Shyh-shiaw Kuo (USA)

Session 1-3-B Classification and Pre-classification
(S) European Language Determination from Image
Takehiro Nakayama and A. Lawrence Spitz (USA)
(S) Classifier Combination for Hand-Printed Digit Recognition
Michael Sabourin, Amar Mitiche, Danny Thomas and George Nagy (CANADA)
(S) Handwritten Numeral Recognition Based on Multiple Feature Extractors
L. Heutte, J. V. Moreau, B. Plessis, J. L. Plangnaud and Y. Lecourtier
(S) On The Bias of Mahalanobis Distance due to Limited Sample Size Effect
Tetsuo Takeshita, Shigeyuki Nozawa and Fumitaka Kimura (JAPAN)
(S) A Structural Indexing Method for Character Recognition
Angelo Marcelli, Natasha Likhareva and Theo Pavlidis (USA)
(S) Two-Stage Box Connectivity Algorithm for Optical Character Recognition
Radovan Krtolica and Sophia Malitsky (USA)
(S) A Zip Code Recognition System Using the Localized Arc Pattern Method
Mitsu Yoshimura, Tatsuro Shimizu and Isao Yoshimura (JAPAN)

Session 1-3-C On-Line Character Recognition
(S) Writer-Adaptation for On-Line Handwritten Character Recognition
N. Matic, I. Guyon, J. Denker and V. Vapnik (USA)
(S) A New Approach to On-Line Handwritten Chinese Character Recognition
Y. Liu, L. Zhang and J. W. Tai (CHINA)
(S) Multiple Word Segmentation with Interactive Look-Up for Cursive Script
R. K. Powalka, N. Sherkat, L. J. Evett and R. J. Whitrow (UK)
(S) A Cursive On-Line Hangul Recognition System Based on the Combination of
Line Segments
Oh-Sung Kwon and Young-Bin Kwon (KOREA)
(S) On-Line Japanese Character Recognition Experiments by an Off-Line Method
Based on Normalization-Cooperated Feature Extraction
Masahiko Hamanaka, Keiji Yamada and Jun Tsukumo (JAPAN)
(S) An On-Line Character Recognition System for Effective Japanese Input
K. Toyokawa, K. Kitamura, S. Katoh, H. Kaneko, N. Itoh and M. Fujita

16:30 - 17:00 Plenary Session
Award Presentations

17:00 - 18:00 Invited Speech
(K) Personal Communicators, Handwriting, Hypertext, and the New Mobility
Norman K. Meyrowitz, Director, System/User Software, GO Corporation (USA)

18:00 - 20:00 Reception

Thursday October 21--------------------------------------------------------
9:00 - 10:45
Session 2-1-A Contextual Processing I
(S) Improving Speed of Japanese OCR through Linguistic Preprocessing
Kenichiro Kigo (JAPAN)
(S) A Post-Processing Method for Handwritten Kanji Name Recognition Using
Furigana Information
K. Marukawa, M. Koga, Y. Shima and H. Fujisawa (JAPAN)
(S) Text Recognition Enhancement with a Probabilistic Lattice Chart Parser
Tao Hong and Jonathan J. Hull (USA)
(L) On Virtual Partitioning of Large Dictionaries for Contextual Post-
Processing to Improve Character Recognition
Rainer Hoch and Thomas Kieninger (GERMANY)
(S) Efficient Postprocessing Algorithms for Error Correction in Hangul
Address and Human Name Recognition
Seong-Whan Lee, Eun Sun Kim and Byung Woo Min (KOREA)
(S) Semantic Analysis for Large Vocabulary Cursive Script Recognition
T. G. Rose, L. J. Evett and R. J. Whitrow (UK)

Session 2-1-B Neural Networks
(I) Connected Character Recognition with a Neural Network
Kunihiko Fukushima (JAPAN)
(L) Optimal Design of Reference Model Using Simulated Annealing Combined with
Improved LVQ3
Hee-Heon Song and Seong-Whan Lee (KOREA)
(S) Printed Japanese Character Recognition Based on Multiple Modified LVQ
Neural Network
Kageyasu Miyahara and Fumio Yoda (JAPAN)
(S) Feedback Pattern Recognition by Inverse Recall Neural Network Model
Keiji Yamada (JAPAN)
(S) Dynamic Handwritten Chinese Signature Verification
Hong-De Chang, Jhing-Fa Wang and Hong-Ming Suen (CHINESE, Taipei)

Session 2-1-C Thinning and Thresholding
(I) Validation of Preprocessing Algorithms: A Methodology and its Applications
to the Design of a Thinning Algorithm for Handwritten Characters
Rejean Plamondon, Marc Bourdeau, Claude Chouinard and Ching Y. Suen
(S) Thresholding Using An Illumination Model
J. R. Parker, C. Jennings and A. G. Salkauskas (CANADA)
(S) Threshold Selection Using Second Derivatives of the Gray Scale Image
Theo Pavlidis (USA)
(S) An Object Attribute Thresholding Algorithm for Document Image Binarization
Ying Liu, Richard Fenrich and Sargur N. Srihari (USA)
(S) A Methodology for The Characterization of the Performance of Thinning
M. Y. Jaisimha, Robert M. Haralick and Dov Dori (USA)
(S) Evaluation of Thinning Algorithms from an OCR Viewpoint
Louisa Lam and Ching Y. Suen (CANADA)

10:45 - 11:00 Break

11:00 - 13:00
Session 2-2-A Cursive Word Recognition
(S) Interpretation of Handwritten Address in US Mail Stream
Sargur N. Srihari, Venu Govindaraju and Ajay Shekhawat (USA)
(S) Using n-Grams for The Definition of a Training Set for Cursive Handwriting
Volkmar Pflug (GERMANY)
(L) Recognition of Handwriting Words in a Limited Dynamic Vocabulary
Michel Gilloux, Jean-Michel Bertille and Manuel Lerox (FRANCE)
(L) Using Constrained Snakes for Feature Spotting in Off-line Cursive Script
Andrew W. Senior and F. Fallside (UK)
(S) Cursive Script Recognition: A Fast Reader Scheme
Didier Guillevic and Ching Y. Suen (CANADA)
(S) Performance Analysis of an OCR System via a Handwritten Character Generator
Cheng-Huang Tung, Yuan-Jong Chen and Hsi-Jian Lee (CHINESE, Taipei)

Session 2-2-B Layout Analysis II
(S) Nested Segmentation: An Approach for Layout Analysis in Document
Xiaolong Hao, Jason T. L. Wang and Peter A. Ng (USA)
(L) Document Image Segmentation and Text Area Ordering
Takashi Saitoh, Michiyoshi Tachikawa and Toshifumi Yamaai (JAPAN)
(L) Address Block Location Using Character Recognition and Address Syntax
Jenchyou Lii, Paul W. Palumbo and Sargur N. Srihari (USA)
(S) Language-Free Layout Analysis
David J. Ittner and Henry S. Baird (USA)
(S) Page Segmentation by Segment Tracing
O. T. Akindele and A. Belaid (FRANCE)
(S) A Method to Extract Templates for Robust Symbol Recognition
C. De Stefano, F. Tortorella and M. Vento (ITALY)

Session 2-2-C Interpretation of Drawings
(S) Recognition of Logic Diagrams by Identifying Loops and Rectilinear
Soo Hyung Kim and Jin Hyung Kim (KOREA)
(L) Rastor-to-Vector Conversion by Line Fitting Based on Contours and Skeletons
Osamu Hori and Satohide Tanigawa (JAPAN)
(L) Adaptable Drawing Interpretation Using Object-Oriented and Constraint-Based
Graphic Specification
Boris Pasternak and Bernd Neumann (GERMANY)
(S) Interpretation of Telephone System Manhole Drawings
Juan F. Arias, Chan Pyng Lai, Surekha Chandran, Rangachar Kasturi and
Atul Chhabra (USA)
(S) A New System for the Analysis of Schematic Diagrams
Anja Habacha Hamada (FRANCE)
(S) Recognition of Dimensions in Engineering Drawings Based on Arrowhead-Match
Weidong Min, Zesheng Tang and Long Tang (CHINA)

13:00 - 14:15 Lunch

14:15 - 14:45 Invited Talk I
(T) A History and Evolution of Document Information Processing
Dr. Toshiyuki Sakai, Dean, Faculty of Science and Technology,
Ryukoku University, Professor Emeritus, Kyoto University (JAPAN)

14:45 - 15:15 Invited Talk II
(T) The Role of Document Analysis and Understanding in Multi-Media Information
Prof. Dr. Andreas Dengel, Scientific Director, German Research Center
for Artificial Intelligence (DFKI) (GERMANY)

15:15 - 15:30 Break
15:30 - 17:30 Poster Session (titles are shown below)
18:30 - 20:30 Banquet

Friday October 22----------------------------------------------------------
9:00 - 10:45
Session 3-1-A Handwritten Character Recognition II
(S) State of The Art of Handwritten Numeral Recognition In Japan: Results of
The First IPTP Character Recognition Competition
Toshihiro Matsui, Ikuo Yamashita, Toru Wakahara and Makoto Yoshimuro
(S) High-Order Statistically Derived Combinations of Geometric Features for
Handprinted Character Recognition
A. K. Chhabra, Z. An, D. Balick, G. Cerf, P. Sheppard, R. Smith and
B. Wittner (USA)
(L) Performance Evaluation of Nonlinear Shape Normalization Methods for the
Recognition of Large-Set Handwritten Oriental Characters
Seong-Whan Lee, Jeong-Seon Park and Yuan Y. Tang (KOREA)
(S) Preprocessing and Feature Extraction for a Handwritten Recognition System
T. Caesar, J. M. Gloger and E. Mandler (GERMANY)
(S) Practical Implementation of A Radial Basis Function Network for Handwritten
Digit Recognition
Bernard Lemarie (FRANCE)

Session 3-1-B Robust Recognition of Printed Documents
(I) From Pixels to Paragraphs: the Use of Models in Text Recognition
Sargur N. Srihari (USA)
(S) Performance Metrics for Printed Document Understanding Systems
J. Kanai, T. A. Nartker, S. V. Rice and G. Nagy (USA)
(S) Optical Recognition of Braille Writing
J. Mennens, L. Van Tichelen, G. Francois and J. Engelen (BELGIUM)
(S) Certifiable Optical Character Recognition
Daniel P. Lopresti and Jonathan S. Sandberg (USA)
(S) Building an Object-Oriented Environment for Document Processing
G. Boccignone, A. Chianese, M. De Santo and A. Picariello (ITALY)

Session 3-1-C Systems and Applications
(S) Robust Recognition of Documents
Joachim Kreich (GERMANY)
(S) Understanding Structured Text Documents by a Model Based Document Analysis
T. A. Bayer (GERMANY)
(L) An Integrated System for Italian Bankcheck Verification
G. Dimauro, S. Impedovo and G. Pirlo (ITALY)
(S) VLSI Implementation for HVRI Algorithm in Pattern Recognition
Yuan Y. Tang, Tao Li and Seong-Whan Lee (CANADA)
(L) Refinement and Testing of a Character Recognition System Based on Feature
Extraction in Grayscale Space
Bill Sakoda, Jiangying Zhou and Theo Pavlidis (USA)

10:45 - 11:00 Break

11:00 - 13:00
Session 3-2-A Databases for OCR
(S) Handwritten Korean Character Image Database PE92
Dae-Hwan Kim, Young-Sup Hwang, Sang-Tae Park, Eun-Jung Kim,
Sang-Hoon Paek and Sung-Yang Bang (KOREA)
(S) Sorting The Quality of Handwritten Chinese Characters for Setting Up a
Research Database
Sheng-Lin Chou and Shiaw-Shian Yu (CHINESE, Taipei)
(L) CD-ROM Document Database Standard
I. T. Phillips, S. Chen and R. M. Haralick (USA)
(S) The Implementation Methodology for the CD-ROM English Document Database
I. T. Phillips, J. Ha, R. M. Haralick and D. Dori (USA)
(L) Interpreting Word Recognition Decisions with a Document Database Graph
Jonathan J. Hull and Yanhong Li (USA)
(S) Recognition of Handprinted Digits Using Optimal Bounded Error Matching
Thomas M. Breuel (SWITZERLAND)

Session 3-2-B Analysis of Forms, Equations and Tables
(S) The Processing of Form Documents
David S. Doermann and Azriel Rosenfeld (USA)
(S) Understanding Mathematical Expressions in a Printed Document
Hsi-Jian Lee and Min-Chou Lee (CHINESE, Taipei)
(S) Anatomy of a Form Reader
Stephen W. Lam, Ladan Javanbakht and Sargur N. Srihari (USA)
(L) Toward a Practical Document Understanding of Table-Form Documents: Its
Framework and Knowledge Representation
Toyohide Watanabe, Qin Luo and Noboru Sugie (JAPAN)
(S) Interpretation of Tables
Surekha Chandran and Rangachar Kasturi (USA)
(S) Multiple Combined Recognition System for Processing of Credit Card Slip
Jonghyun Paik, Sungyun Jung and Yillbyung Lee (KOREA)
(S) An Interactive Document Image Description for OCR of Handwritten Forms
David Monger, Graham Leedham and Andy Downton (UK)

Session 3-2-C Hypermedia and Content Retrieval
(S) Using Electronic Facsimiles of Document for Automatic Reconstruction of
Underlying Hypertext Structures
A. Myka and U. Guentzer (GERMANY)
(S) A Collaborative Supporting Method between Document Processing and Hypertext
Shin'ichi Satoh, Atsuhiro Takasu and Eishi Katsura (JAPAN)
(S) Retrieval of Document Images Using Layout Knowledge
Per Herrmann (GERMANY)
(L) Document Image Retrieval System Using Character Candidates Generated by
Character Recognition Process
Shuji Senda, Michihiko Minoh and Katsuo Ikeda (JAPAN)
(L) Learning of Personal Visual Impression for Image Database Systems
Takio Kurita and Toshikazu Kato (JAPAN)
(S) Radiographs as Medical Documents: Automating Standard Measurements
Dov Dori, Adam I. Harris, Gustavo Gambach and Robert M. Haralick (ISRAEL)

13:00 - 14:15 Lunch

14:15 - 15:45
Session 3-3-A Segmentation of Touching Characters
(S) A Projection Reducing the Constraint of Direction
Gerard Lambert (FRANCE)
(S) A Method for Recognizing Character Strings in Maps Using Linguistic
Akira Nakamura (JAPAN)
(S) Cut Classification for Segmentation
T. A. Bayer and U. Kressel (GERMANY)
(S) Segmentation of Touching Characters in Printed Document Recognition
Su Liang, M. Shridhar, M. Ahmadi (USA)
(S) Word Recognition in a Segmentation-Free Approach to OCR
C. H. Chen and J. L. DeCurtins (USA)
(S) Segmentation of Handwritten Digits Using Contour Features
N. W. Strathy, C. Y. Suen and A. Krzyzak (CANADA)

Session 3-3-B Classification and Pre-classification II
(S) Writing Style Detection by Statistical Combination of Classifiers in Form
Reader Applications
Juergen Franke (GERMANY)
(S) Recognition Enhancement by Linear Tournament Verification
Hiroyasu Takahashi and Thomas D. Griffin (USA)
(S) On Discrimination of Handwritten Similar Kanji Characters by Subspace
Method Using Five Features
M. Ohkura, Y. Shimada, M. Shiono and R. Hashimoto (JAPAN)
(S) Perfect Metrics
Tin Kam Ho and Henry S. Baird (USA)
(S) Combination of Multiple Classifiers with Measurement Values
Y. S. Huang and C. Y. Suen (CANADA)
(S) A Solution to the Problem of Touching and Broken Characters
Jairo Rocha and Theo Pavlidis (USA)

Session 3-3-C Component Recognition in Drawings
(I) Detection of Dimension Sets in Engineering Drawings
Chan Pyng Lai and Rangachar Kasturi (USA)
(S) Diagram Understanding Utilizing Natural Language Text
Yuichi Nakamura, Ryo Furukawa and Makoto Nagao (JAPAN)
(S) Don't Tell Mom I'm Doing Document Analysis; She Believes I'm in the
Computer Vision Field
Suzanne Collin, Karl Tombre and Pascal Vaxiviere (FRANCE)
(S) Appropriate Base Representation Using a Run Graph
Gladys Monagan and Markus Roosli (SWITZERLAND)
(S) Optical Recognition of Chemical Graphics
R. Casey, S. Boyer, P. Healey, A. Miller, B. Oudot and K. Zilles (USA)

15:45 - 16:00 Break

16:00 - 17:15
Session 3-4-A Contextual Processing II
(L) A Fast Algorithm for Finding the Nearest Neighbor of a Word in a Dictionary
(S) A Method of Training Dictionaries for Handwritten Character Recognition
Using Relaxation Matching
T. Horiuchi, K. Toraichi, K. Yamamoto and H. Yamada (JAPAN)
(S) A Multi-Classifier Combination Strategy for The Recognition of Handwritten
Cursive Words
B. Plessis, A. Sicsu, L. Heutte, E. Menu, E. Lecolinet, O. Debon and
J. V. Moreau (FRANCE)
(S) Postprocessing Algorithm Based on the Probabilistic and Semantic Method
for Japanese OCR
Akiko Konno (JAPAN)

Session 3-4-B Layout Generation and Decomposition
(S) Automated Acquisition of Rules for Document Understanding
F. Esposito, D. Malerba and G. Semeraro (ITALY)
(S) Text String Extraction within Mixed-Mode Documents
Frank Hoenes and Juergen Lichter (GERMANY)
(S) Extraction of Character Strings from Unformed Document Images
Kei Takizawa, Daisaku Arita, Michihiko Minoh and Katsuo Ikeda (JAPAN)
(S) Intelligent Document Understanding: Understanding Photographs with Captions
Rohini K. Srihari (USA)
(S) Incremental Acquisition of Knowledge about Layout Structures from Examples
of Documents
Koichi Kise, Naoko Yajima and Noboru Babaguchi (JAPAN)

Session 3-4-C Map Interpretation II
(S) Attributes Extraction for French Map Interpretation
J. M. Ogier, J. Labiche, R. Mullot and Y. Lecourtier (FRANCE)
(S) Map Understanding for GIS Data Capture: Algorithms for Road Network Graph
Marc Pierrot Deseilligny, Herve Le Men and George Stamon (FRANCE)
(S) Development of a Map Vectorization Method Involving a Shape Reforming
Naoya Tanaka, Takeshi Kamimura and Jun Tsukumo (JAPAN)
(S) Automatic Generation of Custom Document Image Decoders
Gary E. Kopec and Phil A. Chou (USA)
(S) Parallel Automatic Recognition of Cartography for Constructing Electrical
Distribution Database
Shigeru Shimada, Yoshinori Takahara, Hiroshi Suenaga and Kouji Tomita

17:15 Conference Ends

[Poster Session]
> Block Segmentation and Text Area Extraction of Vertically/Horizontally
Written Document
Naohiro Amamoto (JAPAN)
> A Hybrid Page Segmentation Method
Masayuki Okamoto and Makoto Takahashi (JAPAN)
> A Labeling Approach for Mixed Document
A. Belaid and O. T. Akindele (FRANCE)
> Classification of Compound Document Image Patterns by MCR Stroke Index
A. B. C. Zidouri, S. Chinveeraphan and M. Sato (JAPAN)
> Weak Model-Dependent Page Segmentation and Skew Correction for Processing
Document Images
John F. Cullen and Koichi Ejiri (USA)
> Page Grammars and Page Parsing: A Syntactic Approach to Document Layout
Alan Conway (IRELAND)
> Table Structure Recognition Based on Textblock Arrangement and Ruled Line
Katsuhiko Itonori (JAPAN)
> Image Based Typographic Analysis of Documents
David S. Doermann and Richard Furuta (USA)
> Raster to Object Conversion Aided by Knowledge Based Image Processing
Enrico Nardelli, Michelangelo Fossa and Guido Proietti (ITALY)
> Layout analysis
Johann Wieser and Axel Pinz (AUSTRIA)
> A Functional Classification Approach to Layout Analysis of Document Image
Kazumi Iwane, Masaki Yamaoka and Osamu Iwaki (USA)
> Cursivewriter: On-Line Cursive Writing Recognition System
Peng Zhao, Toshihiro Yasuda and Yukio Sato (JAPAN)
> Real Time Signature Recognition: A Method for Personal Identification
Giuseppe Gazzolo and Luigi Bruzzone (ITALY)
> On-Line Text/Drawings Segmentation of Handwritten Patterns
Kimiyoshi Machii, Hidehiro Fukushima and Masaki Nakagawa (JAPAN)
> A Contour Fill Method for Alpha-Numeric Character Image Generation
K. Nakashima, M. Koga, K. Marukawa, Y. Shima and Y. Nakano (JAPAN)
> A Color Classification Algorithm
Kansei Iwata and Gabriel Marcu (JAPAN)
> Global and Local Document Degradation Models
Tapas Kanungo, Robert M. Haralick and Ihsin Phillips (USA)
> Visual Interface for Retrieval of Electronic-Formed Books
Toyohide Watanabe, Yuichi Kanai and Noboru Sugie (JAPAN)
> Co-Operative Work in a Hypertext System for Universities
Claudiu Popescu (ROMANIA)
> Fast Image Filtering Implementation
M. Shaban Mohammed, S. Lorenzo, L. Nozal and R. M. Boncho (SPAIN)
> Automatic Generation of Chinese Character Structure Data Based on
Extracting the Strokes
Jinghui Wang and Shinji Ozawa (JAPAN)
> Stamped Character Recognition Method Using Range Image
M. Inoh, T. Horiuchi, K. Toraichi, H. Yamada and K. Yamamoto (JAPAN)
> Automatic Seal Identification Using Fluency Function
Y. Yang, T. Horiuchi and K. Toraichi (JAPAN)
> A New Method for The Segmentation of Mixed Handprinted Chinese/English
Hsing-Hung Kuo and Jhing-Fa Wang (CHINESE, Taipei)
> Segmentation and Recognition for Handwritten 2-Letter State Names
S. Tsuruoka, F. Kimura, Y. Miyake, M. Shridhar and Z. Chen (JAPAN)
> Recognition and Representation of Text Characters Using Rubber Band
Jingzhi Jiang, Woontae Kim and Hideyoshi Tominaga (JAPAN)
> Recognition of Handwritten Digits by Consensus Among Independent Learning
Vector Quantization
Tin Kam Ho (USA)
> Utilization of Large Disordered Sample Sets for Classifier Adaptation
Complex Domains
T. Caesar, J. M. Gloger and E. Mandler (GERMANY)
> Multifont Chinese Character Recognition Using Side-Stroke-End Feature
Gang-Bai (CHINA)
> Optical Recognition of Handwritten Chinese Characters by Partial Matching
An-Bang Wang, Jun S. Huang and Kuo-Chin Fan (CHINESE, Taipei)
> A Model Based Detecting Approach for Feature Extraction of Off-Line
Handwritten Chinese Character Recognition
J. W. Tai, Yin-Jian Liu and Li-Qin Zhang (CHINA)
> Script Representation by a Generalized Skeleton
J.-C. Pettier and J. Camillerapp (FRANCE)
> Comparison of Methods for Detecting Corner Points from Digital Curves -
A Preliminary Report,
Keiichi Abe, Rika Morii, Kaori Nishida and Tokuhisa Kadonaga (JAPAN)
> The Feasibility of a Parallel Processing Oriented Character Recognition
Method Based on a Dynamic Model
Rodney G. Webster and Masaki Nakagawa (JAPAN)
> Effect of the Perturbed Correlation Method for Optical Character
Michio Yasuda, Kazuhiko Yamamoto and Hiromitsu Yamada (JAPAN)
> 2-Stage Character Recognition by Detection and Correction of
Error-Identified Characters,
Cheng-Huang Tung and Hsi-Jian Lee (CHINESE, Taipei)
> Relationship Between the Construction of Chinese Character and the Correct
Ratio of Writer Identification
Yoko Seki and Noriyoshi Takasawa (JAPAN)
> On Using Syntactic Constraints in Text Recognition
R. M. K. Sinha (INDIA)
> From Engineering Drawings to 3-D CAD Models: Are We Ready Now?
Dov Dori and Karl Tombre (FRANCE)
> A Multipurpose Drawing Understanding System with Flexible Object-Oriented
Wei Wu and Masao Sakauchi (JAPAN)
> Logo Recognition Using Geometric Invariants
David Doermann, Ehud Rivlin and Isaac Weiss (USA)
> A Color Map Processing System PU-CMPS
Qinghuai Gao, Jun He, Guisheng Qiu and Qingyun Shi (CHINA)
> An Object-Oriented Kernel for Geographical Information Systems
Ying Ming Lin, Takeshi Sagara, Yutaka Ohsawa and Masao Sakauchi (JAPAN)
> Printed Music Recognition
R. Randriamahefa, J-P. Cocquerez, C. Fluhr, F. Pepin
and S. Philipp (FRANCE)
> Machine Visualization, Understanding and Interpretation of Polyhedral
Line-Drawing in Document Analysis,
P. S. P. Wang (USA)
> A Study on a Method of Dividing Machine-Parts into Function Group for
Technical Illustrations
Inho Lee, Seiichiro Dan, Tadahiro Kitahashi and Norihiro Abe (JAPAN)
> An Improved Classification for Cursive Script Recognition
Laurent Duneau and Bernadette Dorizzi (FRANCE)
> Pattern Analysis and Evaluation of Printed Circuit Boards
Masayasu Ito, Yoshinori Takeuchi, Isao Fujita and Tooru Uchida (JAPAN)
> A Robust Detector for Music Staves
Ivan Leplumey, Jean Camillerapp, and Guy Lorette (FRANCE)
> Musical Score Recognition: A Hierarchical and Recursive Approach
Jean-Pierre Armand (FRANCE)
> A New Algorithm Separating Text String from Map Images
Tong Xu and Xinggang Lin (CHINA)
> Towards Automatic Evaluation of Drawing Analysis Performance -
A Statistical Model of Cadastral Map
Dariusz Madej and Andrej Sokolowski (TURKEY)
> Symbols Recognition and Reconstruction Surface from Topographic Map by
Parallel Method
Kazuhiko Yamamoto, Hiromitsu Yamada and Shigeru Muraki (JAPAN)
> Retrieval Method of Textile Pictures Database Using a Complexity Scale
Toru Inohara, Kunio Kondo, Hisashi Satoh and Shizuo Shimada (JAPAN)
> Totally Unconstrained Handwritten Numeral Recognition via Fuzzy Graphs
I. S. I. Abuhaiba and P. Ahmed (SAUDI ARABIA)
> Intelligent Interactive Map Recognition Using Neural-Networks
Hiroyuki Arai, Shinji Abe and Masakazu Nagura (JAPAN)
> Identifying Ability of a Recognition Method Based on the Field of
Michihiro Nagaishi (JAPAN)
> Segmentation and Classification Using Neural Network for Mixed Text/Image
Shinichi Imade, Seiji Tatsuta and Toshiaki Wada (JAPAN)
> Multilayer Perceptron and Uppercase Hand-written Characters Recognition
Gosselin Bernard (BELGIUM)
> A Tree Structured Neural Network
Hazem Raafat and Mohsen A. A. Rashwan (CANADA)
> Some Ergonomic Improvements of Text Error Detection and Prevention in
Victor Backmutsky and Vladimir Zmudikov (ISRAEL)
> A Symbol Recognition System
David Y. Y. Yun (USA)
> Principles of Pen Interface Design for Creative Work
Masaki Nakagawa, Naoki Kato, Kimiyoshi Machii and Toshio Souya (JAPAN)
> An OCR System for Business Cards
Hisashi Saiga, Yasuhisa Nakamura, Yoshihiro Kitamura
and Toshiaki Morita (JAPAN)
> A Contextual Processing for an O.C.R System, Based on Pattern Learning
F. Lebourgeois and J. L. Henry (FRANCE)
> Handprinted Character Recognition System Using Artificial Neural Networks
Adnan Amin (AUSTRALIA)

[Ground Transportations]
The location, Tsukuba Science City, is about 60 km northeast of metropolitan
Tokyo and 40 km northwest of the New Tokyo International Airport (Narita
International Airport).
There is no direct transportation from Narita Airport to Tsukuba. It is not
recommended to take a taxi from Narita Airport to Tsukuba because it costs
more than 20,000 yen(about $200). Please go to Tsukuba by way of JR
(Japan Rail) Tokyo Station described below.

>From Narita Airport to JR Tokyo Station----------------------------
You may take an airport limousine bus or a JR Narita Express (NEX) which runs
directly from Narita Airport to JR Tokyo Station. It takes about 80 minutes by
the limousine bus and about one hour by NEX. The fare is 2,600 yen for the
limousine bus and 2,890 yen for NEX.
You may also take a local train of JR Sobu line instead of NEX. It runs more
often but takes more time, 80 minutes. It costs less expensive 1,260 yen.

>From JR Tokyo Station to Tsukuba-----------------------------------
At the south part of the Yaesu side of JR Tokyo Station, there is Yaesu
Highway Bus Station (The Yaesu side is the east side of the JR station).
There, you may buy a ticket for JR/Kantetsu Highway Bus bound for Tsukuba
Center. It takes about 70 minutes to Tsukuba Center which is the terminal.
The venue of the conference, Tsukuba Dai-ichi Hotel, is only three minute
walk from the terminal. The fare of the highway bus is 1,230 yen.

More about the Highway Bus-----------------------------------------
The running time of the highway bus from Tokyo Station to Tsukuba Center
sometimes becomes much longer than the usual 70 minutes depending on the
The highway bus for Tsukuba Center starts from the bus stop number 2 of Yaesu
Highway Bus Stops. Please be sure not to take a bus bound for Tsukuba-san
(Mt. Tsukuba) which starts from the same bus stop number 2.
As for the bus from Tsukuba to Tokyo, it should be noted that it goes to JR
Ueno Station rather than JR Tokyo Station. It usually takes about two hours
>from Tsukuba to Ueno. The fare is 1,230 yen. Take a train to go to JR Tokyo
Station from JR Ueno Station, which takes only ten minutes.

Please type or print the registration form neatly and send it to the
conference secretary no later than September 20, 1993. The registrants will
receive a confirmation letter which should be presented at the registration
desk to receive conference material. On-site registration is also available
on October 19 and 20 .

Conference Fee-----------------------------------------------------
Members of IAPR, members of its affiliate organizations, and members of
organizations that are supporting this conference are given a discount in the
conference fee as shown below. To encourage students to participate, a special
discount is also available for full-time students. They will be requested to
show their identification at the registration to get the discount.

Table 1 Conference fee
Early/Late registration Before Sept. 1 After Aug. 31
Member 25,000 yen 30,000 yen
Non-member 30,000 yen 35,000 yen
Full-time student 5,000 yen 10,000 yen
Proceedings only 10,000 yen 10,000 yen

Banquet, Tours and CD-ROMs-----------------------------------------
The banquet to be held on Thursday, October 21 requires reservation and
7,000 yen per person. Those who have a food limitation may indicate it in the
registration form. Special menu will be available. The technical tour and
other social tours require reservations although they are free of charge.
Please indicate number of persons in the registration form.
Those who want to receive a commemorative database CD-ROM need to indicate
his/her selection of the four kinds of collections in the registration form.
They will be requested to fill a separate application form on the site. In
the case of students, the form requires a signature of a responsible person
of his/her research group.

Hotel Accommodations-----------------------------------------------
Hotel rooms for about 340 persons are available. The room rate including the
service charge is between 6,200 yen and 10,000 yen for a single room, and
between 12,700 yen and 19,000 yen for a twin room. The tax is separate.
The reserved hotels are grouped into three classes based on the price and the
distance from the conference venue as in the hotel reservation form. Please
indicate your preference in the form and return it to the address shown on
the form. You will receive a confirmation letter with detailed information.
The payments can be made at a hotel front. No deposit is required.
Student Housing for 100 people is also available which costs 3,000 yen per
night. Most of the student house rooms have a single bed without a bath room.
Toilets and bath rooms are for common use. The gate of this house closes at
23:00. The payment need to be made in advance together with the conference
fee. Please see the registration form. Please set up your schedule so as to
show yourself at the registration before using the student house.
The hotel reservation form need to reach Japan Travel Bureau, Tsukuba Branch
no later than September 4. Otherwise, the reservation confirmation depends
upon the room availability at the time your request received. Confirmation
for dates other than October 19-23 will be on a space-available basis.

Payments of the Registration Fee-----------------------------------
Payments are accepted in Japanese yen only. Credit cards of Master Card, Visa
and American Express are acceptable. When payment is done through a bank,
please write the date of the remittance, the bank name and the branch name on
the registration form.

Refund of 70% of the payment will be made if a cancellation notice and a
written request for the refund is received by September 20, 1993. No refunds
will be made for cancellations received after this date.

Notes on registration and payments---------------------------------
> Those who want to reserve the Student House also need to fill and return
the hotel reservation form. Please indicate the dates you want to reserve.
> In case the Student House has no room, the pre-payment will be fully
refunded at the registration. It will be notified in the confirmation letter
and a hotel reservation will be requested.
> As for the on-site registration, the cash and traveler's checks in Japanese
yen only are accepted. Credit cards will not be accepted there.

Address of Conference Secretary------------------------------------
ICDAR'93 Conference Secretary
c/o Prof. K. Yamamoto, Head of Image Understanding Section
Electrotechnical Laboratory
1-1-4 Umezono, Tsukuba
Ibaraki 305, JAPAN
E-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
Fax: +81-298-58-5949