A multiple blind watermark algorithm based on spatial and wavelet domain

Abstract

A multiple blind watermarking scheme is proposed in this paper, which bases on RGB color image decomposition and embeds watermarks in the spatial domain and wavelet domain of gray images decomposed. This method increases the quantity of watermarks embedded, and does not require the original image or original watermarks when extracting watermarks. With the knowledge of cryptography and scrambling transform and chaos theory applied in the paper, the anti-attack capability of watermarks is improved. The extracting process relies entirely on keys. So the security is enhanced. Experimental results demonstrate that this method has better robustness to the shear attack, salt and pepper noise when embedding watermark in the spatial domain and has better robustness to the white Gaussian noise, JPEG compression, and other attacks when embedding watermark in the wavelet domain.