Steganalysis of LSB matching based on histogram features in grayscale image

In this paper, we consider the problem of LSB matching steganalysis in grayscale images and provide a new detection method. We model the matching embedding as a kind of image degradation with certain additive pulse noise proportional to the embedding rate. We get the restored image by wavelet denoising as the cover imagepsilas estimation. Then we extract the features of 1D statistical grayscale histogram of test images and estimated images to train and classify the cover and stego images with support vector machine. Experimental results show that our method is more reliable than the previous related methods.