Attack-resilient Watermarking in the Haar Wavelet Domain

In this paper a watermarking method based on the Haar wavelet transform is proposed. The hiding procedure is performed in the LL subband of the first level of Haar decomposition of the image. The coefficients resilient to specific attacks are modified in order to obtain a robust embedding scheme. The proposed method is used to define a mask to perform the embedding in the wavelet transform domain. As well known from literature, the transmitted data are subject to many different distortions. The aim of this work is to select the coefficients in the LL subband that are more robust to the most common types of distortions like gaussian noise, rotation, motion, blurring, JPEG and sharpening. Several tests have been carried out to verify the effectiveness of the proposed method.