CAJ | 학술논문

Pevny等人2010年提出一种最先进的高度不可检测的自适应空域隐写算法HUGO(Highly Undetectable Steganography)，其保护了相邻四个像素的一阶统计特性，具有很强的抗检测性能，目前国内外针对HUGO检测缺乏有效的方法。由于用HUGO隐写算法进行隐写嵌入时改变部分纹理特征，本文提出一种新的检测方法，利用局部线性变换得到纹理残差图像，计算共生矩阵得到22130维特征向量，最后使用集成分类器进行分类。实验结果表明：在嵌入率为0.4bpp时，针对BOSSRank图像集，获得平均82.71%的检测率，优于Hugobreaker的80.3%和Guel的76.8%，并在低于0.4bpp嵌入率时，其检测效果有所提高。
Pevny등인2010년제출일충최선진적고도불가검측적자괄응공역은사산법HUGO(Highly Undetectable Steganography)，기보호료상린사개상소적일계통계특성，구유흔강적항검측성능，목전국내외침대HUGO검측결핍유효적방법。유우용HUGO은사산법진행은사감입시개변부분문리특정，본문제출일충신적검측방법，이용국부선성변환득도문리잔차도상，계산공생구진득도22130유특정향량，최후사용집성분류기진행분류。실험결과표명：재감입솔위0.4bpp시，침대BOSSRank도상집，획득평균82.71%적검측솔，우우Hugobreaker적80.3%화Guel적76.8%，병재저우0.4bpp감입솔시，기검측효과유소제고。
Since Pevny put forward a kind of the most advanced Highly Undetectable Steganography in 2010,which approximately preserves the joint distribution of the ifrst-order differences between four neighboring pixels and has a strong resistance to detection performance.Now there is a lack of effective methods in HUGO detection at home and abroad. Steganographic embedding with HUGO changed parts of the textural feature,this paper proposes a new detection method,which is ifrst obtained textural residual image by local linear transformation,to get 22130 dimensionality co-occurrence matrix feature vector,ifnal y using the ensemble classiifer to classify.The experimental results show that when the embedding rate is 0.4 bit per pixel,we obtain a detection rate of 82.71%on average using BOSSRank Image Sets,which is better than 80.3%of Hugobreakers,76.8%of Guel, also in less than 0.4 bit per pixel,the detection effect also improved.