Strengthening Image Security with Unimodular Hill Cipher and Advanced Encryption Standard

Digital data security has become increasingly important with the growth of digital image usage, which often contains sensitive information. This study aims to enhance the security of digital images by combining Unimodular Hill Cipher (UHC) and Advanced Encryption Standard (AES) methods. UHC, a matrix-based symmetric encryption algorithm, is used for image matrix encryption, while AES provides an additional layer of security to protect images against advanced cryptographic attacks. The methodology involves several stages: Preprocessing digital images into matrix form, encrypting matrix blocks using UHC, and applying AES encryption for further protection. The encryption and decryption processes are implemented using Python programming, utilizing libraries such as NumPy, Pillow, and PyCryptodome. Experimental analysis evaluates time efficiency, data integrity, and resistance to cryptographic attacks. The results show that the hybrid UHC-AES method significantly enhances security by randomizing pixel values, as evidenced by entropy analysis, histogram comparison, and correlation evaluation. The entropy values indicate high randomness, and the correlation analysis shows no relationship between the original and encrypted images, ensuring strong encryption. Additionally, the method performs efficiently for medium-sized image files, with encryption times increasing proportionally to key size. Despite its effectiveness, challenges remain, particularly in optimizing the encryption speed for large files and addressing key management vulnerabilities. This study contributes to advancing digital image security by integrating UHC and AES into a robust hybrid cryptographic approach. It opens avenues for future research on developing more efficient algorithms and exploring real-time applications in multimedia data protection.