Digital image watermarking using complex wavelet transform

In this paper a new robust digital image watermarking method based on the Complex Wavelet Transform is presented. For improving its robustness features in the algorithm design the Error Correction Code is used. The technique is performed in spatial domain. The Complex wavelet transform is firstly used to adapt the watermark to the local image activity by using the visual masking. Secondly it is implemented to select the embedding space (embedding channels). The two embedding channels are obtained by selecting third and fourth level of CWT decomposition of the original image and by applying the inverse CWT transform on each level separately. First and second levels of CWT decomposition are avoided due to the low robustness on the compression attacks. Watermark is added to the embedding channels using the additive spread spectrum technique. To compare the proposed algorithm in CWT domain with equivalent one derived in the DWT domain, the same embedding procedure is applied in DWT domain. Robustness of both algorithms on selected attacks like JPEG, JPEG2000 compressions, different filtering attacks etc. is tested and compared by using the benchmark software Checkmark.