US-Turkey Cooperative Research: Steganalysis Techniques for Images and Audio

Polytechnic University Of New York, Brooklyn NY

Investigators

Abstract

0218074 Memon Description: This award is for support of a cooperative project by Professor Nasir Memon, Department of Computer Science, Polytechnic University of New York, Brooklyn, New York, Dr. Bulent Sankur, Department of Electric Engineering, Bogazici University, Istanbul, Turkey and Dr. Ismail Avcibas, Department of Electrical and Electronics Engineering, Uludag University, Bursa, Turkey. They plan to develop new steganalysis techniques for audio and image data. The development of techniques for audio and image steganography and the widespread availability of tools for the same have led to an increased interest in steganalysis techniques for image data. Their approach is based on the hypothesis that hiding a secret message in audio or image data leaves definite artifacts that can be detected using statistical analysis. The steganalyzer is built using multivariate regression on selected audio and image features and a training set of cover and stego images. Scope: The three scientists, N. Memon, I. Avcibas and B. Sankur have already been collaborating on an international project for three years. The project is expected to lead to joint publications and to the development of software to detect hidden messages in audio signals and image files. The steganalysis tools developed would be of significant benefit to the intelligence community for surveillance and monitoring purposes. The project involves visits between the PIs for short periods and visits by one U.S. graduate student and two Turkish graduate students for longer periods. The proposed plan also involves co-advising of all the graduate students working on the project.