Paired sampling and selective ensemble based image steganalysis

摘要：

　　This article introduces a novel ensemble classifier for steganalysis of JPEG images.We put forward four different sampling strategies, and built selective ensemble classifier based on paired sampling strategy and genetic optimization for steganalysis.Experiments show that different sampling strategies can influence detection performance in image steganalysis obviously.In contrast to previous approaches in image steganalysis, proposed method can significantly reduce the error rate in detection performance.

关键词： Information hiding steganalysis selective ensemble paired sample genetic optimization

作者: ZHANG Minqing DI Fuqiang LIU Jia

作者单位: Key Laboratory of Network and Information Security under CAPF, Electronic Department,Engineering University of Chinese Armed Police Force, Xi”an 710086

会议类型: 国内会议

会议名称: 第十二届全国信息隐藏暨多媒体信息安全学术大会

会议地点: 武汉

会议语种:英文

页码: 20-28

在线出版日期: 2015-03-28（万方平台首次上网日期，不代表论文的发表时间）