A Pareto-optimal evolutionary approach of image encryption using coupled map lattice and DNA

Evolutionary algorithms are generally a suitable approach for optimization problems, having more than one conflicting objectives. For many complicated engineering optimization problems, multi-objective formulations are treated as realistic models. The paper presents and implements a Pareto-optimal image encryption algorithm that uses coupled map lattice (CML) chaos function and deoxyribonucleic acid (DNA) combination to encrypt an image. The discussed work uses multi-objective genetic algorithm (MOGA) to get the optimized results. The proposed two-step algorithm uses pseudo-random number generators, the chaotic method CML and DNA to create an initial population of DNA masks in its initial stage. The MOGA is applied in the second stage to obtain the best mask for encrypting the given plain image. The focus is on the generation of Pareto fronts by using the Pareto generation method of multi-objective optimization. The paper evaluates the performance of the implemented work using standard metrics like key sensitivity, secret key space, number of pixel change rate, unified average changed intensity, entropy, histogram and correlation coefficient. It also discusses the impact of using a genetic algorithm that uses more than one fitness function as the objective for encrypting images.