A Robust Security Method for Medical Image Based on 2D Schaffer Map
Suvita Rani Sharma, Manpreet Kaur, Birmohan Singh, Sachin Minocha · 2024
In this work, a methodology has been propounded for the security of the medical images containing Confusion and Diffusion phases in which a 2D Schaffer map is utilized. The confusion phase rearranges the row and column of medical images using the sorted indexes of the chaotic sequences produced by the Schaffer map. During the diffusion phase, the chaotic sequences are utilized to alter the pixel values of the original image through the XOR operation. This phase applies to both the row and column pixel values of the images, enhancing security. The efficacy of the proposed security method has been assessed through histogram analysis, chi-square, and variance test, revealing uniformity in encrypted image pixel values. The correlation analysis shows results closer to zero and an entropy value nearly equal to the ideal 8. Moreover, the encrypted image exhibits lower Peak Signal-to-Noise Ratio (PSNR) and higher Mean Square Error (MSE) values, indicating structural dissimilarity with the original image. The Unified Averaged Changed Intensity (UACI) value exceeds 33.40% and the Number of Changing Pixel Rate (NPCR) value exceeds 99.60%, indicating resistance against differential attacks, while the Structural Similarity Measure (SSIM) values closer to zero for the encrypted image indicates structural dissimilarity. The results affirm that the propounded methodology is secure from entropy, differential, statistical, and brute force attacks.