Image Steganography: Secret Data Concealment and Detection in Digital Images using Least Significant Bit Method

Abstract

Nowadays, the number of data breaches increases surprisingly with the advancement of digital information storage. To increase the safety level of confidential data stored in the digital world, image steganography was proposed to prevent information leakage to intruders. Image steganography is the process of concealing secret information within a digital image. Most image steganography methods have compatibility, security and imperceptibility challenges: limited secret data embedded, brute force secret data extraction and suspicious and perceptible stego image generated. This research had proposed a novel image steganography method to hide confidential information inside a digital image using the Least Significant Bit (LSB) substitution. The usage of a compressed archive file for accepting multiple secret message was to be encrypted. A strong Advanced Encryption Standard (AES) symmetric encryption was combined with the Secure Hash Algorithm-256 (SHA-256) hashing algorithm to provide secret key protection to increase security. Random substitution in LSB increased the imperceptibility of the stego image generated. Secret messages were hidden in the stego image where human eyes cannot detect the difference compared to the cover image. The imperceptibility of the proposed method was evaluated in terms of parameters Peak Sound to Noise Ratio (PSNR) and Structural Similarity Index (SSIM). The overall PSNR results obtained were above 50 dB, where the minimum acceptable value was only 30 dB. Visual comparison and histogram analysis indicated that the changes in the visual structure of the image were very tiny, which will not be more than 0.02 %, as proven in SSIM results. In conclusion, the proposed LSB image steganography is a secure steganography method that accepts multiple types of secret messages to generate an imperceptible stego image with the secret key protection using AES.