Keyword Spotting From Online Chinese Handwritten Documents Using One-Versus-All Character Classification Model

Abstract

In this paper, we propose a method for text-query-based keyword spotting from online Chinese handwritten documents using character classi¯cation model. The similarity between the query word and handwriting is obtained by combining the character classi¯cation scores. The classi¯er is trained by one-versus-all strategy so that it gives high similarity to the target class and low scores to the others. Using character classi¯cation-based word similarity also helps overcome the out-of-vocabulary (OOV) problem. We use a character-synchronous dynamic search algorithm to e±ciently spot the query word in large database. The retrieval performance is further improved by using competing character confusion and writer-adaptive thresholds. Our experimental results on a large handwriting database CASIA-OLHWDB justify the superiority of oneversus-all trained classi¯ers and the bene¯ts of con¯dence transformation, character confusion and adaptive thresholds. Particularly, a one-versus-all trained prototype classi¯er performs as well as a linear support vector machine (SVM) classi¯er, but consumes much less storage of index¯le. The experimental comparison with keyword spotting based on handwritten text recognition also demonstrates the e®ectiveness of the proposed method.