基于自产生投票的手写汉字识别

在模式识别领域,投票策略是非常有效的,而且已被成功应用到人脸检测、识别等领域. 然而,在手写汉字识别 (Handwritten Chinese character recognition, HCCR)中,由于类别集很大、训练样本少等特点,现有的很多分类器集成方法方法都很难直接应用于此领域. 本文提出一种自产生式投票的方法,该方法通过事先学习得到的参数集产生一个测试集合,然后用一个分类器去识别 测试集合中的每个样本,得到属于各个类别的概率,最后通过加权投票得到识别结果. 实验结果表明,本文提出的方法是实用和有效的.

Handwritten Chinese Character Recognition Based on Self-generation Voting

Voting strategy is very useful in pattern recognition and it has been successfully applied in many applications like face detection and recognition. However, these state-of-the-art methods are infeasible or unsuitable for handwritten Chinese character recognition (HCCR) because of the problem's characteristics. In this paper, a self-generation voting based method is proposed for further improving the recognition rate in handwritten Chinese character recognition. This method learns a set of parameters first for generating a set of samples from the test sample, and then classify these generated samples using a base-line classifier. At last, it gives the recognition result by voting. Experimental results on two databases show that the proposed method is effective and useful in handwritten Chinese character recognition systems.