Recognition-based Segmentation of On-Line Cursive Handwriting

Part of Advances in Neural Information Processing Systems 6 (NIPS 1993)

Bibtex Metadata Paper


Nicholas Flann


This paper introduces a new recognition-based segmentation ap(cid:173) proach to recognizing on-line cursive handwriting from a database of 10,000 English words. The original input stream of z, y pen coor(cid:173) dinates is encoded as a sequence of uniform stroke descriptions that are processed by six feed-forward neural-networks, each designed to recognize letters of different sizes. Words are then recognized by performing best-first search over the space of all possible segmen(cid:173) tations. Results demonstrate that the method is effective at both writer dependent recognition (1.7% to 15.5% error rate) and writer independent recognition (5.2% to 31.1% error rate).