Robust zero-watermarking algorithm for medical images based on SIFT and Bandelet-DCT

The gradual improvement of traditional medicare in the cloud has greatly promoted the development of medical enterprise. Meanwhile, problems such as the leakage of patients’ personal information, the theft, and tamper of medical images transmitted in the cloud have become increasingly prominent. To solve the above problems, a novel zero-watermarking algorithm for medical images based on bandelet and discrete cosine transform (Bandelet-DCT) is proposed. First, Scale Invariant Feature Transform (SIFT) is performed on the original medical image for extracting the features as the preprocessing step. Second, the chaotic system tent map is used to encrypt the watermark which contained patients’ information. Then, Bandelet-DCT is applied to extract the visual feature vectors of medical images. Finally, the watermark embedding and extraction are realized by combining zero-watermarking technology and cryptography related technology. The experimental results show that the proposed algorithm has strong robustness and can effectively solve the problem of information leakage. It has a strong anti-attack ability, good robustness, and certain application prospects.