Segmentation Based Steganalysis of Spatial Images Using Local Linear Transform

Most of the traditional image steganalysis techniques are conducted on the entire image and do not take advantage of the content diversity. However, the steganalysis features are affected by image content, and the impact is more serious than embedding. This makes steganalysis to be a classification problem with bigger within-class scatter distances and smaller between-class scatter distances. In this paper, a steganalysis algorithm aimed at spatial steganographic methods which can reduce the differences of image statistical characteristics caused by image content is proposed. The given images are segmented to sub-images according to the texture complexity. Steganalysis features based on local linear transform are separately extracted from each sort of sub-images with the same or close texture complexity to build a classifier. The final steganalysis result is figured out through a weighted fusing process. Experimental results performed on several diverse image databases and circumstances demonstrate that the proposed method exhibits excellent performances.