浅谈计算机辅助书法真伪鉴定  

On Calligraphy Identification Based on the Computer Assistance

作　　者：刘自强[1] 王芳弟[2] 

机构地区：[1]甘肃省中医院经济管理处,甘肃兰州730050 [2]兰州工业学院基础学科部,甘肃兰州730050

出　　处：《兰州工业学院学报》2015年第3期66-72,共7页Journal of Lanzhou Institute of Technology

基　　金：中国铁路总公司科技研究开发计划项目(2013G009-N)

摘　　要：当前书法真伪鉴定主要依靠专家经验性鉴定,不具有普遍性和客观性,且会造成误判.提出了利用模式识别技术进行计算机辅助书法真伪鉴定的方法.首先对书法作品进行自适应阈值预处理,然后利用Gabor滤波器进行特征提取,最后使用Fisher分类器进行真伪鉴定分类.实验选定7位中国历史上知名书法家的作品进行验证,均得到了很好的鉴定效果.The calligraphy identification mainly relies on the experience of experts, which is not objective and may cause misjudgment. According to this problem, the pattern recognition technology by computer is used to as-sist calligraphy identification. Firstly, the calligraphy is processed by pretreatment of adaptive threshold, then features of processed calligraphy are extracted by the Gabor filter, and lastly, fisher classifier is used to identify the calligraphy. The works of seven famous calligraphers in the history are tested in this paper and it turns out that such technology has a good performance.

关 键 词：书法 真伪鉴定 模式识别 GABOR滤波器 FISHER分类器 

分 类 号：TP391.43[自动化与计算机技术—计算机应用技术]