A Wavelet Watermarking Based on HVS and Watermarking Capacity Analysis

摘要

In this paper, a wavelet digital watermarking algorithm based on human visual system and watermarking capacity analysis was proposed. Firstly, watermark was scrambled through Arnold transform. Then, the frequency, texture and luminance masking factor of the improvement just noticed difference threshold model can be exploited to compute the watermark embedding intension in wavelet domain. And, with the watermarking capacity analysis, adaptive wavelet coefficients were chosen to embed the watermark. Finally, the watermark embedding intensions further adjust by watermark attack analysis. Experimental results show that this algorithm can get a watermarked image with perceptual invisibility. Moreover, the algorithm is robust to typical watermark attacks such as: noise adding, image filter, cropping and JPEG compression.