%0 Journal Article
%T 基于拓扑特征的甲骨文识别<br>Oracle-Bone Inscriptions Recognition Based on Topological Features
%A 曲恒熠
%A 吴极
%A 宋英涛
%A 付鹏杰
%A 郭承昊
%J Computer Science and Application
%P 1111-1117
%@ 2161-881X
%D 2019
%I Hans Publishing
%R 10.12677/CSA.2019.96125
%X <div style=\"text-align:justify;\">
	<span style=\"font-size:14px;\">甲骨文是中国的一种古老的象形文字，是中国商朝的古老成熟文字，甲骨文的研究对弘扬中华文化具有深远的意义。甲骨文的许多字形具有原始图画文字的特点，并且甲骨文分为多个时期，每个时期的甲骨文字形都会有所改变，所以不能使用目前常用的汉字识别方法来识别甲骨文，这造成了甲骨文识别研究的停滞不前。本文通过分析甲骨文的拓扑特征，构建拓扑特征点、连通域、亏格等拓扑特征之间的关系，提出了一种能够高准确识别甲骨文的方法。对甲骨文字典上的甲骨文字进行识别测试，实验表明该甲骨文识别方法能准确并且快速地识别甲骨文字。<br />
Oracle-bone inscription is an ancient hieroglyph in China. It is an ancient mature text of the Shang Dynasty in China. The study of Oracle-bone inscription is of far-reaching significance for the promotion of Chinese culture. Oracle-bone inscription is a hieroglyph with traces of original pictorial text; Oracle is divided into multiple periods, and the oracle bone shape of each period will change, so it is not possible to use the commonly used Chinese character recognition method to identify Oracle-bone inscription, which caused the stagnation of Oracle-bone inscription recognition research. In this paper, by analyzing the topological features of Oracle-bone inscription, the relationship between topological features such as topological feature points, connected domains and genus is constructed, and a recognition method for high-accuracy recognition of Oracle is proposed. The Oracle-bone inscription on the Oracle dictionary was identified and tested. The experiment showed that the Oracle recognition method accurately and quickly recognized the oracle bone text.</span> 
</div>
%K 甲骨文，拓扑特征，识别方法<br>Oracle-Bone Inscription
%K Topological Feature
%K Recognition Method
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=30896