Online Handwriting Mongolia Words Recognition Based on Multiple Classifiers

This paper primarily discussed online handwriting recognition methods for Mongolia words which being often used among the Mongolia people in the North China. We introduced the multiple classifiers which were built on different feature sets. Because of the characteristic of the whole body of the Mongolia words, namely connectivity between the characters, thereby the segmentation of Mongolia words is very important. We make use of online and offline information for feature selection. And online feature applied to HMM classifier, offline feature applied to BP neural network and nearest neighbor classifier. Our classification combined all of these three models. Experimental results show that writer-dependent words achieve recognition rates above 95%. And unconstrained words achieve recognition rates about 90%. Recognition rate achieves just to the level of utility.