LSB matching steganalysis based on feature analysis approach

Steganography is the art and science of hiding information inside a cover object. Although we have number of steganographic techniques like text, audio, video, network and image, image steganography is the one of the most popular steganographic technique in communications. In image steganography, there is a technique in which the least significant bit is modified to hide the secret message, known as the least significant bit (LSB) steganography. Several steganalyzers are developed to detect least significant bit (LSB) matching steganography. Least significant bit matching images are still not well detected, especially, at low embedding rate. In this paper, two features are generated, one is the relative feature and the other is the concatenated feature and compared with the features of other two existing methods and experimentally shown that the detection performance is improved while comparing proposed method features with the other two existing method features.