Dongdong Fu, NJIT
Date : November 22, 2006 (Wednesday)
Time : 10:30 AM
Location : ECE 202, NJIT

Abstract

Nowadays digital images have been easily generated and popularly used. With modern image processing tools, digital images can be easily modified without leaving any visual clue. There is an increasing concern to determine whether an image is authentic or not, especially for forensic purpose. Two problems of digital image forensics are addressed in this dissertation: (1) JPEG image steganalysis and (2) image tampering detection.

The aim of steganalysis is to detect the very existence of information hidden by steganography. This dissertation presents a novel steganalysis scheme to effectively attack the JPEG steganographic schemes. The proposed method exploits the correlations between block-DCT coefficients in both intra-block and inter-block senses and Markov empirical transition matrices are used to capture these correlations, resulting in superior steganalytic capability.

Image splicing is one of the most commonly used image tampering schemes. The image splicing process is generally considered as a highly nonlinear and non-stationary process. This dissertation presents a passive image splicing detection scheme based on the recently developed Hilbert-Huang Transform (HHT) which is suitable in analyzing nonlinear and non-stationary signals. A statistical natural image model has been proposed to distinguish natural images from spliced images, which is based on the moments of characteristics functions using wavelet decomposition.

Finally, motivated by Benford’s law that has been widely used in financial forgery detection, the investigation of applying the Benford’s law to image forensics is conducted in this dissertation research. As a result, a novel statistical model, generalized Benford’s law, for the distribution of the first digits of JPEG coefficients is presented in the dissertation. Some useful image forensic applications of the proposed model have been discussed.

Advisor:

Dr. Yun Q. Shi, Professor, ECE Dept., NJIT

Committee Members:

Dr. Nirwan Ansari, Professor, ECE Dept., NJIT

Dr. Swades K. De, Assistant Professor, ECE Dept., NJIT

Dr. Atam Dhawan, Professor, ECE Dept., NJIT

Dr. Frank Y. Shih, Professor, CS Dept., NJIT

Dr. Edward K. Wong, Associate Professor, CIS Dept., Polytechnic University