Preface

This volume contains papers selected for presentation at the 5th IAPR Workshop on Document Analysis Systems (DAS 2002) in Princeton, NJ on August 19–21, 2002. As such, it represents the contributions of an international community of academic, industrial, and government researchers and reflects the state of the art in such diverse domains as character recognition and classifier design, text segmentation, table understanding, page layout analysis, and document engineering, indexing, and retrieval, along with emerging application areas including the World-Wide Web, where algorithms developed for “traditional” paper documents take their place alongside completely new techniques. DAS 2002 continues in the fine tradition of past workshops held in Kaiserslautern, Germany (1994), Malvern, PA (1996), Nagano, Japan (1998), and Rio de Janeiro, Brazil (2000). DAS is distinguished from other gatherings of document analysis researchers by its emphasis on systems. To extract useful information from a noisy page image is a complex process requiring multiple stages, each of which is a research topic in its own right. The integration of such components introduces additional complications. Hence, building and testing document analysis systems is a challenging task for which no definitive solution yet exists. You will find that the papers in this volume carry with them a “flavor” that is unique to DAS. The 44 regular and 14 short papers that appear herein derive from 15 different countries, highlighting the international make-up of our discipline. Regular papers were subjected to rigorous review by members of the Program Committee as well as other noted researchers. Short papers were chosen by the Co-Chairs for their relevance to the themes of the workshop. We would like to thank the reviewers for their diligence, the members of the Program Committee and past DAS chairs for their support, and Larry Spitz, in particular, for his encouragement. We also wish to acknowledge the generous financial assistance of Avaya Labs, as well as that of Yasuaki Nakano, Co-Chair of DAS 1998. Lastly, we offer our gratitude to Princeton University for making its first-rate facilities available to DAS 2002.