Image Encryption Algorithm Based on Improved Hill Cipher Using the 2D Logistic Map

This paper proposes a novel image encryption method, which is based on an improvement of the classic HILL technique using the 2D logistic map. First vectorize the original image and subdivide it into blocks of three pixels. Then several dynamic vectors are created from the 2D logistics map. These vectors are used with pseudo-random translation vectors, for the implementation of several affine transformations to overcome the problem of the linearity of the classical method of Hill. After that, chaining functions will be used to boost the avalanche effect's impact and defend the new system against any differential attack. Simulations on a set of images demonstrate the good performance of the algorithm in terms of image encryption and different attacks.