Robust Watermarking Algorithm for Medical Volume Data Based on PJFM and 3D-DCT

Abstract

With the rapid development of information technology and digital image processing, digital imaging technology is gradually mature and widely used in the field of medicine. The use of computed tomography (CT) and magnetic resonance imaging (MRI) for diagnosis has become an indispensable and effective auxiliary means of modern medical diagnosis. It not only facilitates the diagnosis of doctors and real-time information exchange, but also effectively solves the problem of uneven distribution of medical resources. However, medical data is vulnerable to various malicious attacks when spreading and sharing information on the network, and there is a risk of disclosure of patient privacy. To solve this problem, this paper proposes a robust watermarking algorithm for medical volume data based on deformed Jacobi-Fourier moments and 3D discrete cosine transform. The algorithm first used PJFM moment to extract the features of each slice of 3D medical volume data. Then the feature matrix was transformed by 3D-DCT to combine the contour information and detail information of each slice, and the visual feature vector was obtained by applying perceptual hash technology. In addition, the proposed algorithm used Logistic Map to preprocess and encrypt the original watermark image, and utilized zero watermarking technology to embed,, which further improved the security and practicability of the algorithm. Simulation results show that the proposed algorithm has good robustness against common and geometric attacks, especially resist rotation attacks, cropping attacks and compression attacks.