Light-Field 3D Image Encryption Using Dynamic Cellular Automata

Abstract

In order to design an efficient light-field three-dimensional (3D) image encryption algorithm with good encryption performance and ensure the security and reliability of the image in the process of information transmission, a light-field 3D image encryption based on computational integral imaging and dynamic cellular automata (CA) is proposed. This method based on the hexagonal lenslet array generation method with higher fill factor generates high quality hexagonal element image array (EIA) by pixel mapping, which can achieve higher spatial resolution and improve the quality of decoded 3D image. Furthermore, compared with the previous 3D image encryption methods, the proposed algorithm greatly increases the key space and improves the security of the algorithm by dividing the EIA image into blocks, and encrypting the light-field 3D image using dynamic CA rules according to the characteristics of each sub image. The effectiveness of the proposed method is tested and verified by experimental results of 3D image encryption and computational reconstruction. The security and encryption efficiency of our proposed approach have been significantly improved through the analysis and test of the key space and sensitivity shown computer simulation analysis and comparison, which has a good application prospect.