Online Handwritten Character Recognition Of New Tai Lue Based On Random Forest

Online handwritten character recognition is an important category of pattern recognition. At present, the Chinese characters, English, Arabic and other widely used digital characters have achieved a relatively mature identification algorithm, The handwritten character recognition New Tai Lue also did not have the releva nt research.Therefore, in order to expand the scope of the handwriting recognitio n algorithm, to contribute to the cause of national culture, the paper online hand written character recognition New Tai Lue do the research. And implemented an on-line handwritten character recognition New Tai Lue WEB system.In this paper, in reference to the characters, Japanese and other text recognit ion method based on the combination of the characteristics of the New Tai Lue handwritten characters, on-line handwriting recognition New Tai Lue handwritten characters were studied, the main work is as follows:1, put forward a set of handwritten character recognition method for New T ai Lue, including preprocessing, feature extraction, classification and recognition.2, random forest classification technology as the core of the classification, r ecognition rate performed better.3, the identification method to make an identification module, and integrated into a WEB recognition system.4, the last work of this paper is summarized, and the future research work carried out some prospects.Experiment has achieved good results, the recognition rate reached 88.4%, to achieve the desired recognition effect, which proves random forest classifier ava ilability on character recognition problem.