Selective region medical image encryption algorithm based on cascade chaos and two-dimensional Joseph traversal

Abstract Aiming at the problem that medical image information is easy to be stolen, tampered and maliciously destroyed in the transmission process, a selective region medical image encryption algorithm based on cascade chaos is proposed. The encryption algorithm consists of two stages: region of interest (ROI) encryption and full image encryption. First, the Cubic map is cascaded with an improved Cosine map. Through the analysis of dynamic characteristics, the cascade map has better chaotic performance than the traditional chaotic systems. The key sequence is generated by the cascade map and the image. Then edge detection is performed on the image to determine ROI. ROI is divided into multiple image blocks, and a new gradient algorithm is used to label the sub-blocks containing important information. A modified two-dimensional Joseph traversal method is used to scramble the positions of pixels in each sub-block, and further modify their pixel values with a round of diffusion. Finally, we use a block V-shape scrambling method to scramble the entire image. The final encrypted image is obtained through a round of bidirectional diffusion of all pixels using chaotic sequences. Simulation results and security analysis show that the algorithm has large key space and sufficient security. It can effectively resist various malicious attacks.