A wavelet-domain watermarking technique based on support vector regression

Abstract

A Wavelet-Domain watermarking technique based on Support Vector Regression(SVR) is proposed in this paper. First, the image is decomposed through wavelet transform. Then we use the relationship between the selected coefficient and its neighboring coefficients to train SVR. Thanks to the good learning ability of SVR, the watermark is adaptively embedded in wavelet domain and also can be extracted by the well trained SVR. Experimental results show that the proposed method's performance is better than the one based on SVR in spatial domain in the high robustness to common image processing and the JPEG compression.