Detection of Double-Compression in JPEG2000 by using Markov Features
The detection of double compression in JPEG2000 can make crucial contribution to the steganalysis. We propose an effective approach based on 2-D Markov chain of thresholded prediction-error image to detect the double compression in JPEG2000. The thresholded prediction-error images are generated by subtracting horizontal, vertical, main diagonal and minor diagonal pixel values from current pixel values respectively and then thresholding with a predefined threshold T. Then we use the Markov random process to model the thresholded prediction-error in order to utilize the second-order statistics. Support vector machine is used as the classifier. Experimental results have shown that our proposed approach has outperformed the prior arts.
Double JPEG2000 Compression Discrete Wavelet Transform Prediction Error Image Markov Support vector machine
Zhao Fan Shilin Wang Shenghong Li Yujin Zhang
Department of Electronic Engineering Shanghai Jiao Tong University, Shanghai, China Department of Information Security Engineering Shanghai Jiao Tong University, Shanghai, China
国际会议
厦门
英文
113-116
2010-10-26（万方平台首次上网日期，不代表论文的发表时间）