Genetic algorithm based imperceptible spatial domain image steganography technique with high payload capacity

Abstract

Data security is a very important factor in any form of digital communication. Steganography can be used to enhance the security of digital communication. There are various methods to perform steganography on digital images, but very few deal with increasing imperceptibility and data embedding capacity together. In this paper, a high data embedding capacity spatial domain image steganography scheme is proposed which is highly imperceptible. In the proposed technique steganography is modeled as a search and optimization problem and Genetic algorithm (GA) is used to solve this problem to find the near-optimal solution. Optimal pixel adjustment procedure (OPAP) is further used to improve the quality of stego-image. Experimental results exhibited that the proposed technique provides an improvement in imperceptibility of stego-image at high data embedding rate when compared to several other popular steganography techniques. The average PSNR value of various stego-images at two bit per pixel data embedding rate was 46.39.