1(College of Computer Science, Shaanxi Normal University, Xi′an 710062, China)2(Shaanxi Key Lab of Speech & Image Information Processing,Xi′an,710072, China)3(School of Science, Ningxia Medical University, Yinchuan 750004, China)4(Information and Navigation Institute, Air Force Engineering University, Xi′an 710077, China)
Abstract：To obtain a better balance of robustness and imperceptibility in digital watermarking, this paper proposes a novel watermarking method in wavelet domain which employs the Cuckoo Search Algorithm. In this method, the third-level horizontal detail sub-image obtained by applying 3-Level Wavelet Transform to the cover image is selected as the embedded position, and the watermark is scrambled with Arnold Transform and changed into a one-dimensional bit stream. Then the Cuckoo Search Algorithm is employed to determine the best embedding coefficient. Both the robustness and imperceptibility are taken into account in designing the objection function of the Cuckoo Search Algorithm, where the former makes a full consideration of the attacks such as compressing, scaling, cutting, filtering, and noise disturbance, and the latter is designed according to the wavelet coefficients of the cover images before and after embedding the watermark. Experimental results show that compared with some existing methods based on swarm intelligence, this method has stronger robustness for anti-attack while preserving the imperceptibility.
