Improved Quantization Watermarking with an Adaptive Quantization Step Size and HVS.

This paper proposes a new image-adaptive watermarking technique which utilizes a new combination of an adaptive quantization step size and a HVS(human visual system) model in the wavelet domain. Here we use Quantization Index Modulation(QIM) method with an adaptive quantization step size to realize the embedding scheme. The HVS masking is accomplished pixel by pixel by take into account the luminance and the frequency content of all the image subbands. The watermarking consists of a pseudorandom sequence which is adaptively embedded into the subbands. As usual, the watermark bits are detected by a minimum distance detector. Experimental results prove the effectiveness of the new algorithm. [ABSTRACT FROM AUTHOR]