计算机专业毕业论文 关于手写汉字的骨架提取的研究.docVIP

关于手写汉字的骨架提取的研究 RESEARCH ABOUT EXTRACTION OF HAND WRITING CHINESE CHARACTER SKELETONIZATION 专 业：网络工程 姓 名： 指导教师姓名： 申请学位级别：学 士 论文提交日期：2007年6月15日 学位授予单位：天津科技大学  摘 要 汉字是历史悠久的中华名族文化的重要结晶，闪烁着中国人民智慧的光芒。汉字是一个非字符化的文字，其多体化、多形态化的特性，尤其是手写体的因人而异的形态特性，使得汉字的识别成为计算机文字处理的最大障碍。汉字识别是解决汉字自动输入的问题。一个高效且识别率高的汉字识别系统，就能解决大量汉字输入难的问题，对于计算机在我国的应用和普及具有十分重大的意义。 汉字的识别首先是识别出笔划及其汉字的大概形态特征，即汉字骨架。汉字骨架提取是数字图像预处理中的重要一环，有关汉字图像的骨架提取算法研究一直受到汉字识别和图像处理领域的广泛关注。本文运用了对汉字从外向内腐蚀的细化算法对汉字的图像进行了骨架提取并提取了笔划特征。实验结果表明：该算法汉字骨架连通性较好，提取的骨架能反映汉字的形态特征。 关键词：骨架；汉字识别 ；图像处理 ；细化  ABSTRACT Chinese characters is a long history of Chinese ethnic culture of crystallization , is glittering the Chinese people wisdom ray. The Chinese character is a non-character writing, its multi-body, many of the characteristics of shape, especially the different shape characteristic of the handwritten form person to person, causes the Chinese character recognition into computer language processing biggest barrier. Chinese character recognition is to solve the problem of automatic input Chinese characters. An efficient and high recognition rate of the Chinese character recognition system, the difficult problems of large number of Chinese character input will be able to solve, the computer application in China and penetration is of great significance. The first of the Chinese character recognition is to identify the broad strokes and morphological characteristics of Chinese characters, namely, Chinese skeletonization. Chinese skeletonization from pretreatment in the digital image is an important part, the Chinese extraction algorithms of the skeletonizatoin image have been on Chinese character recognition and image processing areas of concern. In this paper, the use of Chinese characters within outward from the corrosion of refining the image of Chinese characters algorithm to extract the skeleton from a stroke and characteristics. The results showed that: the skeletonization a