Guangdong Guobi Technology Co. Ltd

Anjin Hu

Binghui Chen

Huaxing Lu

Jingping Li

Chuntao Tan

Fang Guo

Muyu Cai

Honghui Wang

Jing-lian Gao

Zhipin Liu

Zhengwei Wu

Xinchun Huang

Zhiai Wang

The present invention discloses a method for recognizing a handwritten character, which includes the following steps of: obtaining a coarse classification template and a fine classification template; receiving a handwritten character input signal from a user, gathering a discrete coordinate sequence of trajectory points of the inputted character, and pre-processing the discrete coordinate sequence; extracting eigenvalues and calculating a multi-dimensional eigenvector of the inputted character; matching the inputted character with the coarse classification template to select a plurality of the most similar candidate character classes; and matching the eigen-transformed inputted character with sample centers of the candidate character classes selected from the fine classification template, and determining the most similar character classes among the candidate character classes. The present invention further discloses a system for recognizing a handwritten character. The present invention can recognize an inputted character fast at a high recognition precision.

Method and system for recognizing a handwritten character

Growing community of inventors

Guangzhou, China

Anjin Hu