This document traces the development of a new cryptosystem using two circuits ensured by a deep Vigenere classical technique improvement, separated by a genetic operator. This new technique employs several dynamic substitutions matrices attached to chaotic replacement functions; whose construction will be detailed. Firstly, we will be start by modifying the seed pixels by an initial value ​​calculated from the original image, and will be infected through the chaotic map used to overcome the uniform image problem, followed by the improvements Vigenere injection technology. The output vector will be subdivided into sub blocks for future application of deeply improved genetic mutations to better adapt to color and medicals image encryption. The second round will increase the complexity of the attack and improve the installed systems. Simulations performed on a large number of images of different sizes and formats ensure that our approach is not exposed to known attacks.

The pixel of an image is always thought of as being the most important component in the encryption process. Our new technology will act on the bit by treating it as the key component of the encrypted image by using new dynamic replacement functions and strong diffusion functions that take values from tables of random size substitutions manufactured from multiple linear congruence generators that subscribe to the Hull and Dobell theorem. Two algebraic operators combined with two genetic operators will be applied at the DNA level in the output. In order to demonstrate that, with this new technology, our system is more susceptible to all known assaults, a huge number of images selected at random from databases were engaged in the simulations.