Information Audit Based on Image Content Filtering

被引：4

作者：

YU Fei SHEN Yue AN Jiyao ZHANG Lingfen ZHU Miaoliang Institute of Computer and Information Engineering Hunan Agricultural University Changsha Hunan China Institute of Artificial Intelligence Zhejiang University Hangzhou Zhejiang China Institute of Computer and Communication Hunan University Changsha Hunan China ^{[1
,2
,1
,3
,1
,2
,1
,410128
,2
,310027
,3
,410082
]}

机构：

来源：

Wuhan University Journal of Natural Sciences | 2006年 / 01期

关键词：

information audit system; content audit; texture character; character library;

D O I：

暂无

中图分类号：

TP391.41 [];

学科分类号：

080203 ;

摘要：

At present, network information audit system is almost based on text information filtering, but badness information is embedded into image or image file directly by badness information provider, in order to avoid monitored by. The paper realizes an information audit system based on image content filtering. Taking the pornographic program identification for an example, the system can monitor the video including any abnormal human body information by matching the texture characters with those defined in advance, which consist of contrast, energy, correlation measure and entropy character measure and so on.

引用

页码：234 / 238

页数：5