计算机应用研究Issue(1)：255-257,265,4.DOI:10.3969/j.issn.1001-3695.2016.01.059
基于富模型的分层量化隐写分析算法
Layered quantitative steganalysis using rich model
摘要
Abstract
Concerned with the bad effect of the detection of the images with lower embedding rate under the current quantita-tive steganalysis,this paper proposed a method of layered quantitative steganalysis.It initially used loss function consisted of changing weight coefficient related to payload value for testing to detect and estimate the payload range.Then it segmented the results of assessment into several parts,and used the loss function of increasing corresponding piecewise weight for a second de-tection.The experimental results show that,compared with classical gradient boosting tree algorithm,the detection effect of pro-posed algorithm for lower payload image is better improved,and the overall detection has higher accuracy.关键词
量化隐写分析/分层检测法/损失函数/梯度增量/负载值Key words
quantitative steganalysis/layered detection/loss function/gradient boosting/payload分类
信息技术与安全科学引用本文复制引用
张聪,张敏情,刘佳..基于富模型的分层量化隐写分析算法[J].计算机应用研究,2016,(1):255-257,265,4.基金项目
国家自然科学基金资助项目 （）
