中文摘要：

甲骨文作为古文字还没有进入国家标准也没有进入国际标准,所以甲骨文字在出版物上是以图片出现,这给检索带来了困难;同时使用大数据进行甲骨文考释的过程中,需要大量的已标注的甲骨图像数据库,而人工标注耗时耗力,且只有甲骨文专家能够完成这项任务,基于以上两个原因,甲骨文字图片的识别变得越来越重要,本文采用支持向量机分类技术研究甲骨文字图片的识别技术,通过试验证明达到88%的准确率。

英文摘要：

Oracle Bone Inscription Recognition based on SVM LIU Yongge, LIU Guoying （ School of Computer and Information Engineering , AnYang Normal University, AnYang 455000, China） Abstract ：Oracle bone inscription as ancient characters has not yet entered the national standard and the inter- national standard, so the characters in a publication are pictures, which bring difficulty to the retrieval process, at the same time; the interpretation of Oracle bone inscription needs annotated image which annota- tion is time -consuming and only Oracle experts can complete it, based on the above two reasons, the image recognition characters become more and more important, this paper we use support vector machine classifica- tion technology to do research of Oracle bone inscription image recognition, the result of experiment is 88% accuracy accurate. Key words： oracle bone inscription ; SVM ; recognition