Zero-watermarking Algorithm for Medical Volume Data Based on Legendre Chaotic Neural Network and Perceptual Hashing

Abstract

Medical information digitization makes the medical information storage and extraction more convenient. Medical image information security and copyright protection is also gradually being taken seriously, and some medical image watermarking has been applied. According to the characteristics of three-dimensional medical images, this paper proposes a robust zero-watermarking algorithm for medical volume data based on legendre chaotic neural network and perceptual hashing. The algorithm is based on three-dimensional discrete wavelet transform frequency analysis features, which uses perceptual hashing technique to extract medical volume data itself feature vector in order to structure robust zero watermarking. And using legendre chaotic neural network to generate chaotic sequence to enhance the security of the watermarking. The algorithm achieves a combination of legendre chaotic neural network encryption and zerowatermarking technology, which can improve the medical volume data watermarking algorithm security and robustness. The simulation results show the effectiveness of the algorithm.