Statistical steganalyis of images using open source software

In this paper we present a novel steganalytic tool based on statistical pattern recognition. The main aim of our project was to design and implement a system able to classify the images into ones with no hidden message and steganographic images using classic pattern classification techniques such as Bayesian classification and decision trees. Experiments are conducted on a large data set of images to determine the classification algorithm that performs better by comparing classification success and error rates in each case. We have employed Weka, a data-mining tool developed in java for this purpose. We have also developed an application using Weka Java library for loading the data of the Images and classify the images into normal images and steganographic images. This application runs a GUI(Graphical User Interface) that enables the user to choose the classifier and other options required for the classification. Our results are aligned with current state of the art research and have the advantage of using open source software.