Detection of Stego Anomalies in Images Exploiting the Content Independent Statistical Footprints of the Steganograms

Steganography which facilitates covert communication creates a potential problem when misused for planning criminal activities. Its counter measure steganalysis is focused on detecting (the main goal of this research), tracking, extracting, and modifying secret messages transmitted through a subliminal channel. In this paper, a feature classification technique, based on the analysis of content independent statistical properties, is proposed to blindly (i.e., without knowledge of the steganographic schemes) determine the existence of hidden messages in an image. To be effective in class separation, the genetic-X-means classifier was exploited. For performance evaluation, a database composed of 5600 plain and stego images (generated by using seven different embedding schemes) was established. Based on this database, extensive experiments were conducted to prove the feasibility and diversity of our proposed system. Our main results and findings are as follows: 1. a 80%+ positive-detection rate.(promising rate for a blind steganalyzer) 2. The removal of content dependency from features enhances the discriminatory power of the classifier. 3. Universal, blind steganalyzer. (not limited to the detection of a particular steganographic scheme ) 4. Detection of stego images with an embedding rate as low as 5% of the maximum payload.