Secure Image Encryption Using Dynamic Block Segmentation and Adaptive Pixel Modification with Chaotic Masking

The widespread use of digital technologies has resulted in an increased generation, transmission and storage of digital images, which need to be secured effectively and efficiently. This paper proposes and evaluates a novel image encryption algorithm that comprises two secure components as its diffusion and confusion modules, i.e., a dynamic block segmentation and permutation module for diffusion and an adaptive pixel modification technique with chaotic masking for confusion. The dynamic block segmentation and permutation module divides the plain image in variable block sizes and then permutes the blocks randomly based on a chaotic permutation key. The adaptive pixel modification with chaotic masking module, on the other hand, obscures the pixel values of the permuted image in a cascaded manner, which ensures that the modification of one pixel affects all pixels within a block. Extensive security analysis validates the effectiveness of the proposed algorithm by exhibiting an entropy of 7.998 and a correlation value of 0.0004. These results indicate that proposed algorithm produces a cipher image with near-perfect entropy and almost zero correlation, making it highly resistant to cyberattacks.