A novel perceptually adaptive image watermarking scheme by selecting adaptive threshold in dht domain

Abstract

This paper proposed a novel image watermarking technique by applying the characteristics of the human visual system, in Hadamard transform domain. Statistical information measures were used to select proper blocks for data embedding. Watermark was embedded by the modification of Discrete Hadamard transform (DHT) coefficients of selected blocks. Threshold and modification value were selected adaptively for each image block, which improved robustness and transparency. The proposed algorithm was able to withstand a variety of attacks and image processing operations like rotation, cropping, noise addition, resizing, lossy compression and etc. The experimental results showed good performance of the proposed scheme in comparison with some of the recently reported watermarking techniques.

Keywords: Digital image watermarking, Hadamard transform, Entropy, Lossy compression, Adaptive Threshold