摘要： 近年来，网格方向特征已广泛应用于许多手写体汉字识别系统中，并认为是目前较成熟的手写体汉字特征之一。网格技术和方向分解是网格方向特征的两个关键技术。该文提出了一种新的网格技术—扇形网格法，结合边缘方向分解技术[1]，构造了一种新的手写体汉字特征提取方法—扇形网格边缘方向分解特征。实验结果验证了本方法的有效性。

关键词: 特征提取, 手写体汉字识别, 环形网格技术, 方向特征

Abstract: Recently, it is found that the directional feature is considered suitable for chinese character recognition, and it has been widely used as one of the mainstream feature extraction approach. Directional decomposition algorithm and meshing method are two key factors in extraction of directional feature. This paper presents a new elastic circle meshing feature extraction method for handwritten chinese character recognition (HCCR). Experiment result shows the effectiveness of the proposed approach.

中图分类号: 

• TP391.4