Handwriting is a natural medium of recording personal information, in addition to being a medium of communication between individuals. While tremendous advances have been made in computer recognition of speech-which is the most natural means of human communication-recognition of handwriting is still in its infancy.

Although computer recognition of handwriting seems to be a solved problem with the ubiquity of PDAs-where writing is on specialized surfaces (called dynamic handwriting)-recognition of handwriting on paper (or static handwriting) poses numerous challenges. In the scenario where ordinary pen in put to ordinary paper the issues encountered are similar to those in computer vision, e.g., object extraction from background and occlusion, as well as absence of temporal information, having no writing boxes, etc . The best one can hope is to extract relevant information without perfect recognition-by using contextual information to overcome ambiguity, extracting cognitive visual features, learning writer characteristics, etc.

This talk will describe extraordinary success in extracting information from handwriting in specific domains-- postal address recognition (for the USPS) and determining writership (for forensic document examination). More general information extraction principles will be described. A system for searching handwriting databases and extracting writer information will be demonstrated.

Sargur (Hari) Srihari is a SUNY Distinguished Professor at the University at Buffalo, the State University of New York (SUNY), in the Department of Computer Science and Engineering where he teaches courses in machine learning and pattern recognition. He is the founding director of the Center of Excellence for Document Analysis and Recognition (CEDAR)-whose work on handwritten address interpretation led to systems now being deployed by several postal services of the world. More recently he has been involved in the development of information processing models for questioned document examination.

Srihari is a Fellow of the Institution of Electrical and Electronics Engineers (IEEE) and a Fellow of the International Association of Pattern Recognition (IAPR). He is on the Board of Scientific Counselors of the National Library of Medicine, and Technical Advisory Boards of two companies. He received a BE in Electrical Communication Engineering from the Indian Institute of Science, Bangalore, and an M.S and PhD in Computer and Information Science from The Ohio State University, Columbus. He received a Distinguished Alumnus Award by The Ohio State University College of Engineering in 1999 and a best paper award at the International Conference on Document Analysis and Recognition in 2003.