A Secure and High Embedding Capacity Image Steganography Scheme Using Image Enhancement Technique, Compression, and Genetic Algorithm

Data security remains an important aspect of secure data communication. Steganography which hides information within host media such as image and cryptography which scrambles the content of secret message, preventing unauthorized access have become popular techniques for data protection. The goal of steganography is to achieve high embedding payload capacity, robustness against statistical steganalysis attacks and increased security. However, achieving high embedding rate while maintaining enhanced security remains a challenge in steganography. Therefore, the aim of this paper is to propose a technique that achieves higher payload capacity and better corrected pixels to ensure enhanced security protection. An image steganography technique that combines cryptography (ECC), compression, image enhancement technique, genetic algorithm (GA) and implemented in MATLAB® is presented. Secret data is compressed through arithmetic coding and encrypted with secret keys generated using ECC. GA operators (reproduction, mutation, and crossover) are applied to embed the compressed encrypted secret data into host media. Fitness function is calculated to determine best fittest chromosome and retinex-based enhancement is applied to improve visual quality. Experimental results show compression increased the embedding capacity by 26% and significantly increased embedding capacity to 58,487 bits. The method achieved an average PSNR value of 50.16 dB at embedding rate of 4.27 bpp higher than some existing famous studies. An average discrete entropy of 7.9973, SSIM of 0.9729, and standard deviation value of 0.1301 were obtained, indicating a negligible distortion between the cover image and stego image. The proposed method is robust against statistical attacks such as regular singular (RS) steganalysis.