文献类型：期刊文献

英文题名：Attacked image based hidden messages blind detect technique

作者：Mao, Jia-Fa[1,2]; Lin, Jia-Jun[2]; Dai, Meng[3]

机构：[1] Department of Mathematics and Computer Science, Shangrao Normal University, Shangrao 334001, China; [2] College of Information Science and Engineering, East China University of Science and Technology, Shanghai 200237, China; [3] Computer Science and Information Engineering Department, Shanghai Institute of Technology, Shanghai 200233, China

年份：2009

卷号：32

期号：2

起止页码：318

外文期刊名：Jisuanji Xuebao/Chinese Journal of Computers

收录：EI(收录号：20091111955712)

语种：英文

外文关键词：Steganography

摘要：Steganalysis technique is an important part of the security of internet information. It' essential problem is to estimate if the digital carrier takes the secret messages. This paper brings forward the blind detect technique which is based on the hidden messages for aggressive. A 2D feature vector in the spatial and DCT domains is extracted to blindly determine the existence of hidden messages in an image. For evaluation, a database composed of over 7000 cover and stego images (generated by using ten different embedding schemes) was established. Based on the database, simulation experiments were conducted to prove the feasibility and diversity of the proposed system. The proposed system consists of following: (1) The average detection rate is about 90.085%, the total average for electropositive detection rate and negative detection rate; (2) It is not limited to the detection of a particular steganographic scheme; (3) It has the capability to detect the stego images with an lower embedding rate; (4) The test of plain images which is smoothing, sharpening, image size zoomed out, cutting and recompressing can be inspected.