一种基于局部Gabor滤波器组的手写体汉字识别方法

针对单一尺度的Gabor滤波器组只对某一特定粗细的手写体汉字敏感的缺点，提出了一种新颖的多尺度局部Gabor滤波器组。为了评估该方法的识别性能，提出了一个基于Gabor特征的手写体汉字识别系统，实验表明多尺度全局Gabor滤波器组在识别性能上明显提高，局部Gabor滤波器组在基本保持识别性能的情况下，特征维数明显降低，计算量和内存需求减少。该方法的创新之处在于选取局部Gabor滤波器，对863 HCL2000手写体汉字数据库的识别，最高平均识别率达到了92.32%，表明了该方法在手写体汉字识别中的有效性。

Handwritten Chinese Character Recognition Based on Local Gabor Filter Bank

This paper proposed a new local Gabor filter bank with multiple scales to, overcome the disadvantage of the traditional Gabor filter bank with a single scale, which is sensitive to the width variation of handwritten Chinese characters. In order to evaluate the performance of our method, a handwritten Chinese character recognition system based on Gabor feature was implemented. Experimental results show that the method is effective to both dimension reduction and recognition performance. The novelty of the method is to select partial Gabor filter bank with part of m scales and n orientations to extract Gabor feature. The best average recognition rate of 92. 32% was achieved, which indicates this method is suitable for handwritten Chinese character recognition.