Using Stroke- or Character-based Self-organizing Maps in the Recognition of On-line, Connected Cursive Script * Lambert Schomaker NICI/University of Nijmegen, P.O.Box 9104, 6500 HE Nijmegen, The Netherlands. E-mail: Schomaker@nici.kun.nl, Fax: +31-80-615938. Abstract - In this study, comparisons are made between a number of stroke- based and character-based recognizers of connected cursive script. In both approaches a Kohonen self-organizing neural network is used as a feature-vector quantizer. It is found that a ”best match only” character-based recognizer performs better than a ”best match only” stroke-based recognizer at the cost of a substantial increase in computation. However, allowing up to three multiple stroke interpretations yielded a much larger improvement on the performance of the stroke-based recognizer. Within the character-based architecture, a comparison is made between temporal and spatial resampling of characters. No significant differences between these resampling methods were found. Geometrical normalization (orientation, slant) did not significantly improve the recognition. Training sets of more than 500 cursive words appeared to be necessary to yield acceptable performance. Character Recognition On-line Cursive Script Neural Networks Self-organizing Maps Temporal Resampling Spatial Resampling stroke-based vs character-based recognition The automatic recognition of cursive script recently enjoys an increased interest of researchers (1) , especially in view of the development of pen-based notebook computers. Currently appearing commercial systems are limited in the sense that they only allow for the (slower (2) ) entry of isolated handprint characters. Also, the difficulty of recognizing unconstrained, connected cursive script still presents a major challenge to researchers in pattern recognition. In this study, comparisons are made between a number of (variants of) recognizers of connected cursive script * . These systems are under development within a European Esprit project, ”Papyrus”, on cursive * Supported by Esprit grant, project P5204. * the ”mixed handprint and cursive” handwriting is not considered in this study 1