A Novel Chaotic Image Encryption Algorithm Based on Coordinate Descent and SHA-256

In the current digital world, the security of image transmission has gained more importance because of the use of multimedia data in communication, healthcare, and cloud storage. The security of images has not been efficiently addressed by traditional encryption algorithms because of the large size of images. Chaotic systems have been widely used for image encryption due to their sensitivity. However, these chaotic-based image encryption algorithms have some drawbacks in key generation. Moreover, most of the chaotic-based image encryption algorithms use key reuse. This paper proposes a new chaotic image encryption scheme that uses coordinate descent optimization and SHA-256 hashing. First, the input image is modified by adding random pixels. Then, a new key is generated by hashing the modified image. The key generation is unique for each image, even for identical images. Chaotic maps such as SEA and FI-PWLCM are used for generating pseudo-random numbers. Furthermore, the SCD optimization method is employed for improving the statistical behavior of the encrypted image. The performance of the new chaotic image encryption scheme is validated through experiments. The new scheme has shown better security, robustness, and reliability.