图像型垃圾邮件过滤技术研究进展  

A Survey on Key Technologies of Image Spam Filtering

作　　者：李鹏[1] 崔刚[1] 

机构地区：[1]哈尔滨工业大学计算机科学与技术学院,哈尔滨150001

出　　处：《智能计算机与应用》2013年第3期28-32,36,共6页Intelligent Computer and Applications

基　　金：国家自然科学基金(61171193)

摘　　要：近年来,图像型垃圾邮件数量的迅速增长使得传统垃圾邮件过滤系统面临重大挑战,并逐渐成为信息安全领域的研究热点。为了能够快速、有效地滤除图像型垃圾邮件,学者们提出了大量的过滤检测方法。首先简要介绍了图像型垃圾邮件给我国带来的影响;然后结合垃圾邮件图像的特征,对图像型垃圾邮件过滤的主要技术:基于近似特征的过滤、基于图像文本特征的过滤、基于图像浅层特征的过滤等进行了分析;接下来对图像型垃圾邮件数据获取方法进行了介绍;最后对过滤技术的研究方向以及面临的挑战进行了讨论和展望。Recently,the rapid growth of image spam makes traditional spam filters facing more and more challenges,and it also becomes a hot research field of information security.In order to be able to filter image spam effectively and efficiently,scholars have proposed a lot of detection methods.Firstly,this paper introduces the effect of image spam brought to our country.Then,according to the characteristics of spam images,the paper analyzes the main filtering techniques： filtering methods based on near-duplicate detection,filtering methods based on image text features,and filtering methods based on image low-level features,etc.Next,the method of image spam dataset acquisition is introduced.And finally,the research direction and challenges of filtering method are discussed.

关 键 词：图像型垃圾邮件 垃圾邮件图像 特征抽取 邮件过滤 近似复制检测 

分 类 号：TP317.4[自动化与计算机技术—计算机软件与理论]