Zhihua Xia, Xinhui Wang, Xingming Sun and Baowei Wang Steganalysis of least significant bit matching using multi-order differences Security and Communication Networks 7
A learning-based steganalysis method is proposed in this paper. In the training process, feature vectors are extracted from original image set and stego image set with a certain “feature extraction” method. The images are represented by these feature vectors. Then the extracted feature vectors are used to train a “classifier” with a certain classification algorithm such as support vector machine. In the testing process, we first extract the feature vector with the same extraction method from the testing image. Then the classifier is used to judge whether the feature vector is extracted from a stego image or not. Feature extraction is key for learning-based steganalysis. In this paper, we calculate multi-order differences horizontally and vertically. Then co-occurrence matrix is used to model the difference to extract features.
Complete the form below and we will send an e-mail message containing a link to the selected article on your behalf