图片垃圾邮件过滤技术综述  

Survey of image spam filtering technology

作　　者：郑冬冬[1] 宋顺林[1] 

机构地区：[1]江苏大学计算机学院,江苏镇江212013

出　　处：《计算机工程与设计》2010年第1期41-44,共4页Computer Engineering and Design

摘　　要：从图片垃圾邮件的现状着手,通过对图片垃圾邮件的分析,将图片垃圾邮件与文本垃圾邮件之间的不同点进行了对比,并对图片垃圾邮件的特征进行了总结。与此同时,对图片垃圾邮件过滤中常用的一些过滤方法,例如OCR(最优字符识别)以及指纹技术进行了介绍,分析了其优缺点,并结合它们自身的缺点提出了一些建设性看法。最后对最新的反垃圾邮件研究成果作了简略描述,并对垃圾邮件的发展作出了展望。The status of the image spam is described, then the image spam is analyzed, and the difference of image spam is contrasted to the text spam. The feature of the image spam is summarized. At the same time, the common technology methods of image spam filtering are introduced such as OCR （optical character recognition） and dactylogram technology, their advantages and disadvantages are analyzed, and some constructive ideas based on their disadvantages are given. Last it introduces the lastest research production of image spam filtering compendiaryly and gives a expectation for the development of the spam.

关 键 词：图片垃圾邮件 图片特征 电子邮件过滤 指纹技术 最优字符识别 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]