摘要
针对目前色情邮件筛选中只针对邮件的文本内容进行过滤的不足,考虑邮件的文本和附件中所携带的图片信息,提出了一种基于证据理论的色情邮件的筛选方法。新方法分别使用一个色情图片检测器和一个文本分类器分别对邮件的附件及其内容进行初判别,然后将两个检测器的判别结果作为证据,使用D-S理论融合两个检测器的输出得到最后的判别结果。实验表明,该方法能有效地提高色情邮件的识别率。
Because ofthe shortcoming ofpomographic E-mail detection methods which only use the text information of an E-mail, considering E-mail body＇s text information and pictures attached by E-mail, a new effective method for pornographic E_mail detection basedon evidence theory was proposed. Firstly, the text and images extracted from an E mail were fed to text classifier and pornographic image detector respectively for elementary recognition. Then, recognition result came out by fusing the two recognition results from these two different classifiers with evidence theory. Experimental results show that the proposed method is advanced the recognition rate of pornographic E_mai effectly.
出处
《计算机工程与设计》
CSCD
北大核心
2005年第10期2833-2836,共4页
Computer Engineering and Design
关键词
证据理论
图片检测
文本检测
色情邮件
融合
D-S effective theory
image detect
text classifier
pornographic E_mai
fusion
