IWFHR-9 (2004) Proceedings

Ninth International Workshop on Frontiers in Handwriting Recognition

(c)2004 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

Created under the permission of IEEE on December 10, 2004

Table of Contents

  • Forward
  • Organizing Committee
  • Program Committee
  • Reviewers
  • Sponsors
  • Technical Sessions
    • Handwriting Recognition and Shape Analysis
    • Classification Techniques
    • Handwriting Analysis and Gesture Recognition
    • Signature Verification and Writer Identification
    • Recognition of Word and Text
    • Document Analysis and Applications
  • Poster Session
    • History and Future Prospects
    • Segmentation and Preprocessing
    • Feature Extraction and Feature Selection
    • Character Recognition
    • Asian Character Recognition
    • Word Recognition and Linguistic Approaches
    • Signature Verification and Writer Identification
    • Applications


Welcome to the Ninth International Workshop on Frontiers in Handwriting Recognition, held at Hitachi Central Research Laboratory, located in Kokubunji, Tokyo, Japan.

On behalf of the organizing committee, we are very happy to announce that we have received 120 submissions of high-quality papers representing the work of authors from 38 institutions and 24 countries. There will be 100 oral and poster presentations during this 3-day workshop. These presentations will cover classification techniques, shape analysis, gesture recognition, signature verification, writer identification, word and text recognition, and document applications. In addition, we will have the following key speakers.

First, we are honored to have an invited speaker, Prof. S. N. Srihari, State University of New York at Buffalo (USA), who will lecture on "Machine Learning in Questioned Handwriting Examination." Forensic applications have been evolving, and scientific and technological researches have begun sometime ago. Prof. Srihari will present a state-of-the-art forensic handwriting examination.

Second, we welcome distinguished guests-Edward J. Kuebert, US Postal Service (USA); Dr. Gerhard Stönner, Deutsche Post World Net (Germany); Joseph Ulvr, Canada Post Corporation (Canada); Dave Evans, Royal Mail (UK); and Hideo Uchida, Japan Post (Japan). They will join our panel discussion on "The Present and Future of the Postal Automation System: In Quest of More Advanced Recognition Technology." The postal application area has long benefited from the technology that has enabled the reading and interpretation of machine- and handwritten postal addresses. The technological application in the next decade will be discussed from both the user's and provider's perspectives.

The Program Committee chaired by Professors F. Kimura, J.-H. Kim, A. Downton, and R. Sabourin has invited many authors who contributed high-quality papers and performed 360 reviews. The executive committee has made arrangements for the technical tours sponsored by Fujitsu Laboratories, Hitachi, NEC, and Toshiba, as well as other social programs. Microsoft sponsored the TabletPC tutorial, another feature of this workshop. IBM's sponsorship has made it possible for us to provide student discounts. We also thank Hiroyuki Mori of Hitachi for developing a Web-based paper review management system.

We hope you all enjoy the exchange of technical and scientific ideas in our beautiful environs, especially during Japan's most colorful season.

Hiromichi Fujisawa
Guy Lorette
General Chairs
October 1, 2004

Organizing Committee

General Chairs

H. Fujisawa, Central Research Laboratory, Hitachi, Japan
G. Lorette, University of Rennes, France

Honorary Chair

K. Yamamoto, Gifu University, Japan

Program Co-Chairs

F. Kimura, Mie University, Japan
J.-H. Kim, Korea Advanced Institute of Science and Technology, Korea
A. Downton, University of Essex, UK
R. Sabourin, Ecole de Technologie Superieure, Canada

Executive Committee

H. Sako (Chair), Hitachi, Japan
M. Nakagawa (Chair), Tokyo University of Agriculture & Technology, Japan
Y. Ishitani, Toshiba, Japan
M. Koga, Hitachi, Japan
S. Senda, NEC, Japan
H. Tanaka, Fujitsu Laboratories, Japan


C. Y. Suen, Concordia University, Canada
S. N. Srihari, State University of New York at Buffalo, USA

Program Committee

  • A. Amin, University of New South Wales, Australia
  • E. Anquetil, INSA/IRISA, France
  • A. Biem, IBM, USA
  • A. de Souza Britto Jr, PUCPR, Brazil
  • H. Bunke, University of Bern, Switzerland
  • M. Cheriet, Ecole de Technologie Superieure, Canada
  • X. Ding, Tsinghua University, China
  • M. Fairhurst, University of Kent, UK
  • V. Govindaraju, SUNY at Buffalo, USA
  • L. Heutte, University of Rouen, France
  • S. Impedovo, University of Bari, Italy
  • B. Irie, Toshiba, Japan
  • S. Jaeger, Maryland University, USA
  • Y. B. Kwon, Chung-Ang University, Korea
  • L. Lam, The Hong Kong Institute of Education, Hong Kong
  • G. Leedham, Nanyang Technological University, Singapore
  • S.-W. Lee, Korea University, Korea
  • C.-L. Liu, Hitachi, Japan
  • S. Naoi, Fujitsu Laboratories, Japan
  • D. Nishiwaki, NEC, Japan
  • M. Okada, Waseda University, Japan
  • S. Omachi, Tohoku University, Japan
  • U. Pal, Indian Statistical Institute, India
  • M. Perrone, IBM, USA
  • G. Pirlo, University of Bari, Italy
  • G. Ratzlaff, IBM, USA
  • G. Rigoll, Munich University of Technology, Germany
  • N. Sherkat, Nottingham Trent University, UK
  • H. Shimodaira, JAIST, Japan
  • M. Shridhar, University of Michigan-Dearborn, USA
  • J. Snowdon, IBM, USA
  • S. Srihari, State University of New York at Buffalo, USA
  • C. Y. Suen, Concordia University, Canada
  • S. Tsuruoka, Mie University, Japan
  • S. Uchida, Kyushu University, Japan
  • C. Viard-Gaudin, University of Nantes, France
  • N. Vincent, University of Tours, France
  • L. Vuurpijl, University of Nijmegen, Netherlands
  • T. Wakabayashi, Mie University, Japan
  • T. Wakahara, Hosei University, Japan


  • E. Anquetil, France
  • A. Biem, USA
  • D. L. Borges, Brazil
  • A. de Souza Britto Jr, Brazil
  • M. Cheriet, Canada
  • A. Downton, UK
  • C. O. A. Freitas, Brazil
  • H. Fujisawa, Japan
  • V. Govindaraju, USA
  • L. Heutte, France
  • G. Houle, Canada
  • J. Hu, USA
  • B. Irie, Japan
  • K. Ishigaki, Japan
  • Y. Ishitani, Japan
  • S. Jaeger, USA
  • A. Kawamura, Japan
  • F. Kimura, Japan
  • A. L. Koerich, Brazil
  • M. Koga, Japan
  • Y. B. Kwon, Korea
  • L. Lam, China
  • G. Leedham, Singapore
  • C.-L. Liu, Japan
  • G. Lorette, France
  • H. Mizutani, Japan
  • M. Nakagawa, Japan
  • M. Nakai, Japan
  • S. Naoi, Japan
  • D. Nishiwaki, Japan
  • M. Okada, Japan
  • L. E. S. de Oliveira, Brazil
  • S. Omachi, Japan
  • T. Pal, India
  • U. Pal, India
  • T. Paquet, France
  • M. Perrone, USA
  • G. Pirlo, Italy
  • G. Ratzlaff, USA
  • G. Rigoll, Germany
  • R. Sabourin, Canada
  • H. Sako, Japan
  • S. Senda, Japan
  • H. Shimodaira, Japan
  • M. Shridhar, USA
  • J. Snowdon, USA
  • F. Sun, Japan
  • S. Tsuruoka, Japan
  • S. Uchida, Japan
  • C. Viard-Gaudin, France
  • N. Vincent, France
  • L. Vuurpijl, Netherlands
  • T. Wakabayashi, Japan
  • T. Wakahara, Japan


Hitachi, Japan
Fujitsu Laboratories, Japan
NEC, Japan
Toshiba, Japan
Microsoft, USA
International Association for Pattern Recognition

Technical Sessions

Handwriting Recognition and Shape Analysis

Online Character Recognition Using Eigen-Deformations 001

H. Mitoma, S. Uchida, and H. Sakoe

Self-Supervised Adaptation for On-Line Text Recognition 002

L. Oudot, L. Prevost, and A. Moises

Handling Spatial Information in On-Line Handwriting Recognition 002

S. Marukatat and T. Artières

Generative Models and Bayesian Model Comparison for Shape Recognition 004

B. Krishnapuram, C. M. Bishop, and M. Szummer

Modulating Population Granularity for Improved Diagnosis of Developmental Dyspraxia from Dynamic Drawing Analysis 005

S. Hoque, M. C. Fairhurst, and M. A. Razian

Contextual Recognition of Hand-Drawn Diagrams with Conditional Random Fields 006

M. Szummer and Y. Qi

Classification Techniques

Support Vector Machines for Handwritten Numerical String Recognition 007

L. S. Oliveira and R. Sabourin

A Classifier Based on Distance between Test Samples and Average Patterns of Categorical Nearest Neighbors 008

S. Hotta, S. Kiyasu, and S. Miyahara

Classification of Time-Series Data Using a Generative/Discriminative Hybrid 009

K. T. Abou-Moustafa, M. Cheriet, and C. Y. Suen

Speeding Up the Decision Making of Support Vector Classifiers 010

J. Milgram, M. Cheriet, and R. Sabourin

Combination of Three Classifiers with Different Architectures for Handwritten Word Recognition 011

S. Günter and H. Bunke

Normalization Ensemble for Handwritten Character Recognition 012

C.-L. Liu and K. Marukawa

Boosting Driven by Error Free Regions 013

R. Lindwurm and J. Rottland

Unsupervised Feature Selection for Ensemble of Classifiers 014

M. Morita, L. S. Oliveira, and R. Sabourin

Using Informational Confidence Values for Classifier Combination: An Experiment with Combined On-Line/Off-Line Japanese Character Recognition 015

S. Jaeger

A Syntax-Directed Method for Numerical Field Extraction Using Classifier Combination 016

C. Chatelain, L. Heutte, and T. Paquet

Handwriting Analysis and Gesture Recognition

Model Structure Selection and Training Algorithms for an HMM Gesture Recognition System 017

N. Liu, B. C. Lovell, P. J. Kootsookos, and R.I. A. Davis

MagicWand: A Hand-Drawn Gesture Input Device in 3-D Space with Inertial Sensors 018

S.-J. Cho, J. K. Oh, W.-C. Bang, W. Chang, E. Choi, J. Yang, J. Cho, and D. Y. Kim

Inertial Sensor Based Recognition of 3-D Character Gestures with an Ensemble of Classifiers 019

J. K. Oh, S.-J. Cho, W.-C. Bang, W. Chang, E. Choi, J. Yang, J. Cho, and D.Y. Kim

Recovering Dynamic Information from Static Handwritten Images 020

Y. Qiao and M. Yasuhara

A Saliency-Based Multiscale Method for On-Line Cursive Handwriting Shape Description 021

C. De Stefano, M. Garruto, and A. Marcelli

Writer Dependent Online Handwriting Generation with Bayesian Network 022

H. Choi, S. J. Cho, and J. H. Kim

Representation and Annotation of Online Handwritten Data 023

A. S. Bhaskarabhatla, S. Madhavanath, M. N. S. S. K. Pavan Kumar, A. Balasubramanian, and C. V. Jawahar

Distinguishing Text from Graphics in On-Line Handwritten Ink 024

C. M. Bishop, M. Svensén, and G. E. Hinton

On-Line Handwritten Documents Segmentation 025

J. Blanchard and T. Artières

Learning to Parse Hierarchical Lists and Outlines Using Conditional Random Fields 026

M. Ye and P. Viola

Signature Verification and Writer Identification

Learning Strategies and Classification Methods for Off-Line Signature Verification 027

S. N. Srihari, A. Xu, and M. K. Kalera

Using HMM Based Recognizers for Writer Identification and Verification 028

A. Schlapbach and H. Bunke

Ink-Deposition Model: The Relation of Writing and Ink Deposition Processes 029

K. Franke and S. Rose

Recent Advancements in Automatic Signature Verification 030

G. Dimauro, S. Impedovo, M. G. Lucchese, R. Modugno, and G. Pirlo

Automatic Writer Identification Using Fragmented Connected-Component Contours 031

L. Schomaker, M. Bulacu, and K. Franke

ER2: An Intuitive Similarity Measure for On-Line Signature Verification 032

H. Lei, S. Palla, and V. Govindaraju

Handwriting Analysis for Writer Verification 033

A. Bensefia, T. Paquet, and L. Heutte

Recognition of Word and Text

N-Gram Language Models for Offline Handwritten Text Recognition 034

M. Zimmermann and H. Bunke

Handwritten Brazilian Month Recognition: An Analysis of Two NN Architectures and a Rejection Mechanism 035

M. N. Kapp, C. O. De A. Freitas, and R. Sabourin

A New View of the Output from Word Recognition 036

M.-P. Schambach

Comparing Natural and Synthetic Training Data for Off-Line Cursive Handwriting Recognition 037

T. Varga and H. Bunke

Handwritten CAPTCHA: Using the Difference in the Abilities of Humans and Machines in Reading Handwritten Words 038

A. Rusu and V. Govindaraju

Fast Two-Level HMM Decoding Algorithm for Large Vocabulary Handwriting Recognition 039

A. L. Koerich, R Sabourin, and C. Y. Suen

Document Analysis and Applications

Decompose-Threshold Approach to Handwriting Extraction in Degraded Historical Document Images 040

Y. Chen and G. Leedham

Text Line Segmentation in Handwritten Documents Using a Production System 041

S. Nicolas, T. Paquet, and L. Heutte

Improving the Structuring Search Space Method for Accelerating Large Set Character Recognition 042

Y. Yang and M. Nakagawa

An Empirical Study of Statistical Language Models for Contextual Post-Processing of Chinese Script Recognition 043

Y.-X. Li and C. L. Tan

Spiral Recognition Methodology and Its Application for Recognition of Chinese Bank Checks 044

H. Tang, E. Augustin, C. Y. Suen, O. Baret, and M. Cheriet

D-Pen: A Digital Pen System for Public and Business Enterprises 045

N. Furukawa, H. Ikeda, Y. Kato, and H. Sako

Effect of Recognition Errors on Information Retrieval Performance 046

A. Vinciarelli

Document Retrieval System Tolerant of Segmentation Errors of Document Images 047

T. Nagasaki, T. Takahashi, and K. Marukawa

Poster Session

History and Future Prospects

History of the International Workshops on Frontiers in Handwriting Recognition 048

S. Impedovo

Segmentation and Preprocessing

A Recognition-Based System for Segmentation of Touching Handwritten Numeral Strings 049

Y. Lei, C. S. Liu, X. Q. Ding, and Q. Fu

Global Shape Normalization for Handwritten Chinese Character Recognition: A New Method 050

C.-L. Liu and K. Marukawa

Handwriting Segmentation of Unconstrained Oriya Text 051

N. Tripathy and U. Pal

Machine-Printed from Handwritten Text Discrimination 052

E. Kavallieratou, S. Stamatatos, and H Antonopoulou

Automatic Segmentation of Unconstrained Handwritten Numeral Strings 053

J. Sadri, C. Y. Suen, and T. D. Bui

Multi-window Binarization of Camera Image for Document Recognition 054

I.-J. Kim

Local Slant Estimation for Handwritten English Words 055

Y. Ding, W. Ohyama, F. Kimura, and M. Shridhar

Segmentation of Handwritten Numerals by Graph Representation 056

M. Suwa and S. Naoi

Feature Extraction and Feature Selection

Character Image Reconstruction from a Feature Space Using Shape Morphing and Genetic Algorithms 057

C. Iga and T. Wakahara

Extraction of Hybrid Complex Wavelet Features for the Verification of Handwritten Numerals 058

P. Zhang, T. D. Bui, and C. Y. Suen

Experimental Analysis of the Modified Direction Feature for Cursive Character Recognition 059

X. Y. Liu and M. Blumenstein

Evaluation of Feature Sets in the Post Processing of Handwritten Pitman's Shorthand 060

S. M. Htwe, C. Higgins, G. Leedham, and M. Yang

An Optimized Hill Climbing Algorithm for Feature Subset Selection: Evaluation on Handwritten Character Recognition 061

C. M. Nunes, A. de S. Britto Jr, C. A. A. Kaestner, and R. Sabourin

Foreground and Background Information in an HMM-Based Method for Recognition of Isolated Characters and Numeral Strings 062

A. de S. Britto Jr, R. Sabourin, F. Bortolozzi, and C. Y. Suen

A New Series of Rotation Invariant Moments Derived by Lie Transformation Group Theory 063

T. Sakata, R Nishii, T. S. Chin, and R. Sawae

Character Recognition

The Reduction of Memory and the Improvement of Recognition Rate for HMM On-Line Handwriting Recognition 064

D. Funada, D. Muramatsu, and T. Matsumoto

Pattern Recognition by Distributed Coding: Test and Analysis of the Power Space Similarity Method 065

T. Kobayashi and M. Nakagawa

Application of Fuzzy Logic to Online Recognition of Handwritten Symbols 066

J. A. Fitzgerald, F. Geiselbrechtinger, and T. Kechadi

A Generic Approach for On-Line Handwriting Recognition 067

S. Marukatat, T. Artières, and P. Gallinari

Learning HMM Structure for On-line Handwriting Modelization 068

H. Binsztok and T. Artières

Diversity-Performance Relationship in a Handwriting Recognition System Based on Bit-Plane Decomposition 069

S. Chindaro, K. Sirlantzis, M. C. Fairhurst, and S. Hoque

On the Choice of Training Set, Architecture and Combination Rule of Multiple MLP Classifiers for Multiresolution Recognition of Handwritten Characters 070

U. Bhattacharya, S. Vajda, A. Mallick, B. B. Chaudhuri, and A. Belaid

Asian Character Recognition

A Method to Accelerate Writer Adaptation for On-Line Handwriting Recognition of a Large Character Set 071

A. Nakamura

An Off-Line Recognition Method of Handwritten Primitive Manchu Characters Based on Strokes 072

G. Zhang, J. Li, R. He, and A. Wang

Online Handwriting Recognition for Tamil 073

K. H. Aparna, V. Subramanian, M. Kasirajan, G. Vijay Prakash, V. S. Chakravarthy, and S. Madhvanath

Comparison of Elastic Matching Algorithms for Online Tamil Handwritten Character Recognition 074

N. Joshi, G. Sita, A. G. Ramakrishnan, and S. Madhavnath

The Clustering Technique for Thai Handwritten Recognition 075

I. Methasate and S. Sae-tang

Word Recognition and Linguistic Approaches

Stability Measure of Entropy Estimate and Its Application to Language Model Evaluation 076

J. Kim, S. Ryu, and J. H. Kim

Lexicon Organization and String Edit Distance Learning for Lexical Post-Processing in Handwriting Recognition 077

S. Carbonnel and É. Anquetil

Combination of Contextual Information for Handwritten Word Recognition 078

G. Koch, T. Paquet, and L. Heutte

Signature and Lexicon Pruning Techniques 079

S. Palla, H. Lei, and V. Govindaraju

Rejection Strategies for Handwritten Word Recognition 080

A. L. Koerich

An Activation-Verification Model for On-Line Texts Recognition 081

L. Oudot, L. Prevost, and M. Milgram

Signature Verification and Writer Identification

The Repeatability of Signatures 082

R. M. Guest

An Off-Line Signature Verification Method Based on the Questioned Document Expert's Approach and a Neural Network Classifier 083

C. Santos, E. J. R. Justino, F. Bortolozzi, and R. Sabourin

Selection of Points for On-Line Signature Comparison 084

M. Wirotius, J.-Y. Ramel, and N. Vincent

An Effective Writer Verification Algorithm Using Negative Samples 085

X. Wang and X. Ding


Generation and Analysis of Handwriting Script with the Beta-Elliptic Model 086

H. Bezine, A. M. Alimi, and N. Sherkat

A Fast HMM Algorithm Based on Stroke Lengths for On-Line Recognition of Handwritten Music Scores 087

Y. Mitobe, H. Miyao, and M. Maruyama

Handwritten Address Interpretation System Allowing for Non-use of Postal Codes and Omission of Address Elements 088

T. Akiyama, D. Nishiwaki, E. Ishidera, K. Kondoh, M. Hayashi, and T. Yamauchi

Handwriting-Based Learning Materials on a Tablet PC: A Prototype and Its Practical Studies in an Elementary School 089

N. Iwayama, K. Akiyama, H. Tanaka, H. Tamura, and K. Ishigaki

Handwritten Chinese Address Recognition 090

C. Wang, Y. Hotta, M. Suwa, and S. Naoi

A Search Method for On-Line Handwritten Text Employing Writing-Box-Free Handwriting Recognition 091

H. Oda, A. Kitadai, M. Onuma, and M. Nakagawa

Base Color Recognition by Tetragonal Regression for Overlapped Watercolors 092

T. Terai, S. Mizuno, and M. Okada

PATRAM-A Handwritten Word Processor for Indian Languages 093

K. Madduri, K. H. Aparna, and V. S. Chakravarthy

The WANDAML Markup Language for Digital Document Annotation 094

K. Franke, I. Guyon, L. Schomaker, and L. Vuurpijl

Recognition and Grouping of Handwritten Text in Diagrams and Equations 095

M. Shilman, P. Viola, and K. Chellapilla

Use of Chatroom Abbreviations and Shorthand Symbols in Pen Computing 096

W. B. Huber, S.-H. Cha, C. C. Tappert, and V. L. Hanson

A System towards Indian Postal Automation 097

K. Roy, S. Vajda, U. Pal, and B. B. Chaudhuri

Verifying the UNIPEN Devset 098

L. Vuurpijl, R. Niels, M. van Erp, L. Schomaker, and G. Ratzlaff

A Study on Decision Rule for Japanese Dictation Test 099

M. Shi, W. Ohyama, T. Wakabayashi, and F. Kimura

Mode Detection and Incremental Recognition 100

S. Rossignol, D. Willems, A. Neumann, and L. Vuurpijl

End of Program