基于数学形态学的图像汉字笔划细化和提取  

Chinese characters stroke thinning and extraction based on mathematical morphology

作　　者：王建平[1] 钱自拓[1] 王金玲[1] 罗国军[1] 

机构地区：[1]合肥工业大学电气与自动化工程学院,安徽合肥230009

出　　处：《合肥工业大学学报（自然科学版）》2005年第11期1431-1435,共5页Journal of Hefei University of Technology：Natural Science

摘　　要：文章提出了一种基于数学形态学的图像汉字笔划细化和提取方法。根据汉字的结构特点,定义了一组新的结构元素序列,保持了原样本图像汉字的连通性和拓扑性的骨架,有效地抑制样本细化后汉字笔划的扭曲和歪斜;在此细化的结果基础上,定义了一种新的汉字笔划提取算法,准确提取出原汉字的横竖撇捺笔划的形状、形态及形式结构信息,取得了理想的效果。In this paper, a novel method for Chinese characters stroke thinning and extraction based on mathematical morphology is proposed. According to the structure of Chinese characters, a new group of structure element sequences is defined so that the connectivity and topologic features of the skeleton of original Chinese characters can be held, and distortion and deflection of the Chinese characters strokes after sample＇s thinning can be restrained effectively. Based on this result of Chinese characters stroke thinning, a new algorithm of stroke extraction method for Chinese characters is defined,then the horizontal, vertical, right-diagonal and left-diagonal strokes of original Chinese characters are extracted exactly,and the information of their shape and structure is acquired.

关 键 词：数学形态学 图像汉字细化 汉字笔划提取 

分 类 号：TP391.43[自动化与计算机技术—计算机应用技术]