Optimization the chaos based image encryption using TLBO algorithm

Image encryption is a crucial component of data security nowadays. Due to its inherent properties such as sensitivity to initial condition, ergodicity and systematic parametrization chaos systems have been extensively used in image encryption. However, one needs to optimize the chaos for each image being encrypted to provide high quality encryption. Features such as high entropy or low pixel correlation are measures of interest in image encryption. In this paper, we propose a novel technique to optimize a chaos based image encryption algorithm. The image is confused using logistic map as chaos function and diffused by modified Knuth shuffling algorithm. We use the logistic map parameters as encryption key. We then harness TLBO and GSA evolutionary algorithms to speed up the optimization process of these parameters. The optimization method yields parameters that lead to lowest correlation among adjacent pixels or highest entropy. We compare the quality and security of our cryptosystem with some famous image encryption methods in the literature.