Abstract
An adaptive blind watermarking algorithm based on image content and ridgelet transform is proposed. Firstly, the image is divided into three different feature regions: smooth, edge and texture. And then, due to the sparse representation of ridgelet transform for image features, the positions of important vision information where watermark should be embedded can be found out. The watermark is embedded in the visual significant ridgelet coefficients, and the watermarking strength is depended on the weight factor beta which is adaptively changed with different image features. The proposed algorithm obtains good trade-off between robustness and invisibility. Experimental results show that our algorithm is robust against attacks such as JPEG, noise, filtering and random crop.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call