Character Recognition : SSPR ' 90 Working Group Report

This report summarizes the discussions of the Working Group on Character Recognition of the IAPR 1990 Workshop on Syntactic and Structural Pattern Recognition, Murray Hill, NJ, 13-15 June 1990. The participants were: H. Baird, T. Bayer, H. Fuji~awa , T. K. Ho, J. Hull, T. Itagaki, D. Lee, S. Liebowitz, O. Matan, G. Nagy, T. Pavlidis, and S. Srihari. George Nagy moderated the discussion and Thomas Bayer prepared this report based on notes by Nagy, Jonathan Hull, and himself. . It was not easy to agree that any problem in this field is definitely solved. Perhaps this is one: on a small number of known fonts of the printed Latin alphabet, at body text sizes (;::: 8 point), under moderate distortions, and in a controlled environment, it should be possible to achieve better than 99.9% top-choice accuracy using a variety of well-unc1"erstood techniques including dictionary context. However, it's important not to forget that even a 99.99% recognition rate is still unacceptably low for many applications. It was considerably easier to draw up a list of open problems.