小篆文字的自动识别

摘要

Xiaozhuan is the language used Qin unified China,and is a major milestone in the development of Chinese characters. In calli-graphy,inscriptions,stone carvings,it is abound. However,due to large differences with the modern Chinese characters,most people can-not recognize these Xiaozhuan text. An automatic identification method of Xiaozhuan text by using computer technology is presented. First,the standard and database for Xiaozhuan font is built. Secondly,the Xiaozhuan text image which users want to identify is scaled to the standard size,and then the image is matched with these characters in the standard database of Xiaozhuan font by iterative closest point algorithm ( ICP algorithm) . Finally,the similarity is computed and the highest similarity ward is selected,that is the output result of rec-ognition. Therefore automatic recognition of Xiaozhuan font is achieved. A lot of experiments have shown this method is effective.%小篆是秦统一后使用的文字，是汉字发展的一个重大里程碑。在书法、碑文、石刻等有大量存在。但是由于与现代汉字差异较大，大多数人无法辨识这些小篆文字。文中提出了一种利用计算机对小篆文字自动辨识的方法。首先构建标准以及小篆字体数据库，然后将用户需要识别的小篆文字图片缩放至标准大小，随后采用迭代最近点算法( ICP算法)与库中的小篆文字进行匹配，最后计算其相似度，而得到的相似度最高文字，也就是识别的输出结果，从而实现小篆字体的自动识别。该方法经过大量实验证明是有效的。