Steganalysis for Calibrated and Lower Embedded Uncalibrated Images

Abstract

The objective of steganalysis is to detect messages hidden in a cover images, such as digital images. The ultimate goal of a steganalyst is to extract and decipher the secret message. In this paper, we present a powerful new blind steganalytic scheme that can reliably detect hidden data with a relatively small embedding rate in JPEG images as well as using a technique known as calibration. This would increase the success rate of steganalysis by detecting data in transform domain. This scheme is feature based in the sense that features that are sensitive to embedding changes are being employed as means of steganalysis. The features are extracted in DCT domain. DCT domain features have extended DCT features and Markovian features merged together in calibration technique to eliminate the drawbacks of both(inter and intra block dependency) respectively. For the lesser embedding rate, the features are considered separately to evolve a better classification rate. The blind steganalytic technique has a broad spectrum of analyzing different embedding techniques The feature set contains 274 features by merging both DCT features and Markovian features. The extracted features are being fed to a classifier which helps to distinguish between a cover and stego image. Support Vector Machine is used as classifier here.