SECURE IMAGE STEGANOGRAPHY USING LEAST SIGNIFICANT BIT AND EDWARD-CURVE DIGITAL SIGNATURE ALGORITHM

Abstract

Image steganography is the process of incorporating text into photos so that the Human Visual System (HVS) cannot detect it and that only the sender and recipient are aware of its presence. There has been numerous works done on image steganography, however some of the challenges outlined by most of the work include low Peak Signal-to-Noise Ratio (PSNR) and low compression rate. This research proposes a model for image steganography that makes use of Edward-Curve Digital Signature Algorithm (EdDSA), Least Significant Bit (LSB), and Diffie-Hellman algorithm. The information to be hidden in the image is first of all encrypted and signed using EdDSA and then the encrypted message is then encrypted using the Diffie-Hellman algorithm to deal with the issue of key exchange. The encrypted message is then embedded in the image using LSB method which is one of the most used steganographic methods. In order to strengthen the security of the information, the model not only conceals information in a picture but also encrypts it and offers a hash. Both subjective and objective analysis are used to assess the aforementioned technique. The experimental results were evaluated based on PSNR and Space Saving and the proposed method outperforms a number of other methods by having higher Peak Signal-to-Noise Ratio (PSNR) values and a high a compression rate. The result of the study was tremendous having a PSNR values of 96.21415, 99.76348, 88.85138, 98.15750 for Baboon, House, Pepper, Dog stego images respectively. The result also had a compression rate of 90%, 42%, 95% and 83% for Baboon, House, Pepper, and Dog respectively. The proposed image steganography model, integrating EdDSA, LSB, and the Diffie-Hellman algorithm, not only effectively conceals information within images but also enhances security through encryption and hashing. The achieved superior performance in terms of higher PSNR values and substantial compression rates, as validated through comprehensive analyses, underscores the efficacy and robustness of the proposed method in safeguarding sensitive information during transmission.