Sorting qualities of handwritten Chinese characters for setting up a research database

Abstract

Setting up a database containing a large volume of character samples is important for research on character recognition. As the character samples are collected, the samples must be classified into the correct category, it is also necessary to establish an index for measuring the quality of the samples so that one can give information about their quality when certain samples are mentioned. An automatic sorting procedure for Chinese characters is proposed based on the Bayes rule. Three features about Chinese characters are used for the sorting. The sorting results are checked by examining curves of the recognition rates for all samples as the samples are arranged with or without sorting. The experimental results show that the sorting method is effective.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>