Home
    Call for Papers
    Papers Submission
    Important Dates
    Committees
    Technical Program
    Invited Speakers
    Workshops
    Tutorials
    Competition
    Awards
    
    Venue
    Accomodation
    Social Program
    Travel Information
    Sponsors
    Links
    Contact
     

 

 
 

Tutorials

ICDAR 2007 TUTORIAL ANNOUNCEMENT
Sunday, September 23, 2007
Grand Hotel Rayon
Curitiba

The organizing committee invites you for a series of tutorials offered by eminent experts in the field for the benefit of the attendees, including graduate students, and attendees wishing to increase their knowledge of contemporary topics related to document understanding and interpretation. The workshops and tutorials will be held at the Grand Hotel Rayon in Curitiba. The following tutorials are being offered on Sunday, September 23, 2007. For additional details, click on the title of the tutorial.


NOTE: The tutorials listed below will only be offered if there is a minimum enrollment of twelve for each tutorial. Further details about cost etc. will be available shortly. For further information please contact Tutorials Chair: M. Shridhar, University of Michigan-Dearborn, USA ()

TUTORIAL 1: Sunday, September 23, 2007, 8:00 am – 12:00 noon
FORENSIC DOCUMENT EXAMINERS APPROACH TO HANDWRITING COMPARISONS / IDENTIFICATION
 by
Marc Gaudreau, Forensic Sciences Division, Canada Border Services Agency.
Samiah Ibrahim, Forensic Sciences Division, Canada Border Services Agency.

Forensic document examination is a diverse field that includes the analysis and comparison of information that appears in many forms on documents. Topics for the tutorial include: history of the discipline; class versus individual characteristics; variation; statistical approaches; genuineness versus spuriousness; opinions and conclusions. This tutorial will serve the following objectives:
 
  • Provide instruction on established practices and methodologies – specifically forensic document examination approach to handwriting and signature identification.
  • Motivate and explain an ICDAR topic of emerging importance – to broaden the scope of both disciplines.

Marc Gaudreau Manager, Forensic Sciences Division, Canada Border Services Agency
Marc is presently manager of the Forensic Document Section of the Canada Border Services Agency (CBSA) with responsibility for overseeing the national delivery of forensic document examination expertise to clients from within the CBSA, to other federal government departments, provincial and local governments, and certain international government bodies. The CBSA team of ink dating specialists, lead by Marc, has been recently nominated to receive the Government of Canada Award for Technical Excellence. He is a member of the Canadian Society of Forensic Sciences, a member of the American Board of Questioned Document Examiners, and the American Society of Questioned Document Examiners.

Samiah Ibrahim Forensic Document Examiner, Canada Border Services Agency
Samiah is presently a forensic scientist with the Forensic Document Section of the Canada Border Services Agency (CBSA). In this function she is responsible for conducting forensic analysis of questioned document cases involving multiple facets of examination including, but not limited to: handwriting, signatures, typewriters, impact and non-impact printing devices, photocopiers, decipherment of latent impressions, detection and decipherment of obliterations, commercial and security printing processes, wet and dry cachets, and examination of official documents (e.g., passports, birth certificates, driver’s licenses). She is an active member of the American Society of Questioned Document Examiners.

 

TUTORIAL 2: Sunday, September 23, 2007, 1:00 pm – 5:00 pm
COMPUTATIONAL FORENSICS: Techniques, Applications and Challenges in the Computed-Based Analysis of Questioned Documents
 by
Katrin Franke, Norwegian Information Security Lab, Gjøvik, Norway.
Elisa van den Heuvel, Netherlands Forensic Institute, Netherlands.

The aim of this tutorial is to lay the foundations and to encourage further discussions on the development of computational methods for the forensic analysis of questioned documents. Researchers and practitioners from the field of computer science will be introduced to the special demands, the procedures and the techniques applied in forensic casework. Questioned documents studied cover ID- and travel documents, bank checks, contracts, last wills and tax declarations. Also, the tutorial exposes current forensic challenges demanding experts developing next generation equipment and tools. The principal objective of this tutorial is to provide a solid basis for researchers own computational method designs. Not only the feasibility of computer science methods as toolset for forensic investigations is discussed, but new computational algorithms are reviewed in order to study the impact of these technologies. In this way participants will learn to design their own approach and to apply it in the appropriate manner.

Katrin Franke is presently an associate professor at the Norwegian Information Security laboratory in Gjøvik, Norway. She obtained her Ph.D. degree at the Artificial Intelligence Institute, University of Groningen, The Netherlands in 2005. She also holds a degree in electrical engineering from the Technical University of Dresden, Germany (1994) and a banker’s diploma (1989). After finishing her studies in 1994, Mrs. Franke began to conduct research at the Fraunhofer IPK in Berlin, Germany. Mrs. Franke focused on digital image processing and pattern recognition for computer based document and signature analysis. Until December 2006, she worked as a project manager at the Department of Security Technology, in charge of founded research and industrial projects on document processing, signature analysis and custom-stamp analysis, applied in banking and in forensics. Mrs. Franke has published several scientific journal articles, peer-reviewed conference articles and edited books. She is a member of international program committees and is actively involved in the organization of international conferences.

Elisa van den Heuvel started in 2005 as scientific forensic handwriting expert in training at Netherlands Forensic Institute. Her scientific background lies primarily in human motor control with special emphasis on handwriting/signature movement execution. In chronology, she received her M.A. from Radboud University of Nijmegen in 1996, and her PhD at Free University of Amsterdam (FUA) in 2000. During her PhD research she also worked at Motor Control Lab, Arizona State University, Tempe, USA.

 

TUTORIAL 3: Sunday, September 23, 2007, 8:00 am – 12:00 noon
MARKOV MODELS FOR HANDWRITING RECOGNITION
by
Gernot A. Fink, University of Dortmund, Robotics Research Institute, Intelligent Systems Group.
Thomas Plotz, University of Dortmund, Robotics Research Institute, Intelligent Systems Group.

The use of Hidden Markov Models (HMMs) for recognizing machine printed or handwritten text is motivated by the success of this method in the field of automatic speech recognition. In contrast to classical approaches to character recognition that rely on a combination of explicit segmentation and classification steps, HMMs allow for a classification of characters or words with implicit segmentation. Therefore, the approach is especially attractive for handwriting recognition where reliable character segmentation is extremely hard to achieve. However, HMMs are not able to describe long distance dependencies as they occur within texts, e.g., at the word level. For this purpose, so-called language models realized as Markov chain models are suited extremely well. Combining both model types in an overall statistical framework allows for the integrated solution of challenging recognition problems. In this tutorial the theoretical concepts behind HMMs and n-gram language models will be presented. For both techniques the mathematical foundations, the algorithms used for estimating the models’ parameters from sample data, and the techniques for applying them to solving segmentation and classification problems will be described.

Gernot A. Fink received the diploma in computer science from the University of Erlangen-N¨urnberg, Erlangen, Germany, in 1991 and the Ph.D. degree (Dr.-Ing.) also in computer science from Bielefeld University, Germany, in 1995. In 2002 he received the venia legendi (Habilitation) in Applied Computer Science from Bielefeld University. From 1991 to 2005 he was with the Applied Computer Science Group at the Faculty of Technology of Bielefeld University. Since 2005 he is professor for Pattern Recognition in Embedded Systems within the Department of Computer Science at the University of Dortmund, Germany, where he also heads the Intelligent Systems group at the Robotics Research Institute (IRF).

Thomas Plotz studied computer science at the University of Cooperative Education, Mosbach, Germany, and at the University of Bielefeld, Germany. He received degrees in computer science in 1998 (Dipl.-Ing.), 2001 (diploma), and 2005 (Ph.D. – Dr.-Ing.), respectively. From 2001 to 2006 he was with the Applied Computer Science Group at the Faculty of Technology of Bielefeld University. In 2006 he joined the Intelligent Systems Group at the Robotics Research Institute of the University of Dortmund, Germany where he holds a post-doctoral research position.

 

TUTORIAL 4: Sunday, September 23, 2007, 1:00 pm – 5:00 pm
DOCUMENT IMAGE COMPRESSION AND COMPRESSED DOMAIN DOCUMENT PROCESSING
by
Utpal Garain, Indian Statistical Institute, Kolkata, India

With the advent of digital libraries, research in document image compression and compressed domain processing finds a new dimension. Document image compression serves an obvious practical purpose: it conserves storage in a database of document images by removing redundancy, it enhances utilization of available bandwidth in a network and it also improves computer systems performance by reducing accesses to storage devices. The first half of the tutorial would, therefore, cover the basics of data compression and then revisits the popular and frequently used algorithms for compression of scan-digitized document images. International standards like CCITT-IV, JBIG, and JBIG2 will be presented with some reasonable details. The role of symbolic compression in JBIG2 will be highlighted. Idea of residue coding will then be introduced and lossy and lossless versions of a JBIG2 compliant method are to be discussed. Apart from achieving high compression, designing of algorithms now-a-days needs to address another important aspect namely, processing of documents in compressed domain. With increasing amount of document images being stored in compressed format, efficient information retrieval in the compressed domain has also become a major challenge for proficient designing of digital libraries. Being able to randomly access and partially decode the compressed data is highly desirable for efficient retrieval. The second half of the tutorial will focus on this issue namely, compressed domain document processing. The discussion on the compressed domain pattern matching and retrieval will broadly be based on JBIG2 compression method. Requirements for designing a complete compressed domain document image retrieval system will be outlined. The existing research in the related area will be surveyed and then the prevailing studies on accomplishing several IR (information retrieval) tasks for document images will be reviewed. Several applications will then be discussed and research needed to realize these applications in the compressed domain would be presented. The application list will contain (i) stop word spotting, (ii) word stemming, (iii) summarization, (iv) keyword spotting, (v) duplicate detection, etc.

Utpal Garain received his bachelor and master degrees in computer science and engineering in 1994 and 1997, respectively from Jadavpur University, India and Ph.D. degree from Indian Statistical Institute in 2005. Dr. Garain started his career in software industry and later on, he joined Indian Statistical Institute where he is, at present, serving as an Assistant Professor. His research interest includes document image analysis including document compression and OCRs, technology development in Indian languages, machine learning, etc. Dr. Garain is one of the key scientists who first developed an Indic script (Devanagari-Hindi and Bengali) OCR system which was because of its high level performance, transferred to industry for commercialization. In the area of DIA, Dr. Garain has so far published more than forty research papers in international journals and conferences of high repute. He has been invited to serve in the program committee of several conferences including ACM-SAC’07, ICDAR’07, etc. Dr. Garain has been regularly reviewing papers for several international journals including IJDAR, IEEE SMC-B, Pattern Recognition, etc. For his significant contribution in pattern recognition and its applications for DIA and language engineering, Dr. Garain received prestigious Young Engineer Award in 2006 from the Indian National Academy of Engineering (INAE).






CALL FOR TUTORIAL PROPOSALS

ICDAR2007 Organizing Committee invites proposals for the Tutorial Program, to be held immediately before to the technical conference (September 23-26). ICDAR2007 Tutorials should serve one or more of the following objectives:

  • Introduce students and newcomers to major topics of ICDAR research.
  • Provide instruction on established practices and methodologies.
  • Survey a mature area of ICDAR research and/or practice.
  • Motivate and explain an ICDAR topic of emerging importance.
  • Introduce expert non-specialists to an ICDAR subarea

NOTE: If you have an idea for a tutorial that does not quite fit the above objectives, please e-mail Dr. M. Shridhar (tutorial chair) atumich edu for a preliminary evaluation.

Proposals should be from two to four pages in length, and should contain the following information:

  • A brief description of the tutorial, suitable for inclusion in the conference registration brochure.
  • A detailed outline of the tutorial, including preferred length (1/2 or 1 day).
  • Characterization of the potential target audience for the tutorial, including prerequisite knowledge.
  • A description of why the tutorial topic would be of interest to a substantial part of the ICDAR audience.
  • A brief resume of the presenter(s), which should include name, postal address, phone and fax numbers, e-mail address, background in the tutorial area, any available example of work in the area (ideally, a published tutorial-level article on the subject), evidence of teaching experience (including references that address the proposer's presentation skills), and evidence of scholarship in the area.
  • The name and e-mail address of the corresponding presenter. The corresponding presenter should be available for e-mail correspondence during the evaluation process, in the case clarifications and discussions on the scope and content of the proposal are needed.

The evaluation of the proposal will take into account its general interest for ICDAR attendees (e.g., a tutorial on object-oriented inheritance will be hardly accepted), the quality of the proposal (e.g., a tutorial that simply lists a set of concepts without any apparent rationale behind them will not be approved) as well as the expertise and skills of the presenters. We emphasize that the primary criteria for evaluation will be whether a proposal is interesting, well-structured, and motivated, rather than the perceived experience/standing of the proposer.

Last but not least, the tutorial should attract a meaningful audience.Those submitting a proposal should keep in mind that tutorials are intended to provide an overview of the field; they should present reasonably well established information in a balanced way. Tutorials should not be used to advocate a single avenue of research, nor should they promote a product.

All proposals should be submitted by electronic mail to the Tutorial Chair, Dr. M. Shridhar , no later than January 31, 2007. Feedback, comments and/or suggestions would be provided within two weeks of receiving the proposal. Final acceptance (or rejection) would be decided by February 28, 2007.

 

 

 

 

 

 

 

Last update 19 August, 2007