Improved detection for robust image watermarking

In a previous paper, the author proposed a robust watermarking method for still images that embeds the binary watermark into the detail subbands of the image wavelet transform. The perceptually significant coefficients are selected for each subband using a different threshold. For greater invisibility of the mark, the approximation subband is left unmodified. The watermark is embedded several times in each subband to achieve robustness. Here, we propose a new type of detection and we test the performance against different types of attacks (lossy compression, AWGN, scaling, cropping, intensity adjustment, filtering and collusion attack).