Abstract:
Digital watermarking.is a technique proposed for multimedia security applications like
content protection and authentication. In digital watermarking, the watermark is embedded
into a multimedia data (cover data), in such a way that distortion of the cover data due to
watermarking is almost imperceptible. In addition, in reversible watermarking the cover data
is restored after the watermark extraction. Reversible watermarking finds widespread use in
military and medical applications, where distortion—free recovery of the original data after
watermark extraction is of utmost importance. This thesis provides an overview of the state-ofthe-
art reversible watermarking techniques. The most popular approach in reversible
watermarking is based on the histogram modification techniques. The present reversible
watermarking algorithms do not exploit the human visual system (HVS) properties effectively.
We propose two algorithms for reversible watermarking based on histogram modification, which
use better HVS models during embedding. The first method is a texture based method, where
JND computation is redefined to include the texture information. The second is surround pixel
prediction based method. It uses all surround pixels in a non-causal window for predicting the
value of a pixel. The dissertation describes the proposed methods after explaining some
histogram modification based methods. Simulation results show that the proposed algorithms
have better performance over previous histogram modification based reversible watermarking
techniques.
