A new encryption technique for multicolor images is outlined in this study. After vectorization of the clear image, an initialization value will be calculated. This value, allows changing only the first pixel. This pixel will start the encryption process. In parallel, in a first step, we will make a confusion by chaotic vectors entirely controlled by another chaotic vector. In a second step, three chaotic substitution matrices of size (256, 256) will be generated. The passage of each pixel through these matrices will also be controlled by another chaotic vector. Finally, a strong link will be set up between the new pixel state and the previous ones to set up better diffusion/confusion. This step increases the avalanche effect. Testing performed through our algorithm on standard images shows the durability of our system.
In recent years, the rapid growth of road traffic density generates a rising request for tools that can be used to analyze and control the traffic networks. The Microscopic traffic simulation is one of the major tools used in the analysis of traffic systems; it provides a very detailed study of the interaction between the elements of the traffic network. Thus microscopic traffic simulation has become an ever increasing field of research and development.The main aim if this paper is to provide a new model for microscopic traffic simulation; Traditional traffic simulation models neglect some real-life factors that need to be considered, such as the effect of random distribution in the entry of lane. This paper combines the Multi-Agent Systems (MAS) and the stochastic process to model the randomness of vehicles arrival at the entry of the lane.The second contribution of this paper is about the internal structure of mobile agents[8] which initially reacts according to the instructions of the Main agent (MA); in the case of a lack of dynamic information, the mobile agents take decisions based on their experiences accumulated during previous interactions.The obtained results illustrate that using the randomness in the reaction of agent enhanced greatly the performance of simulation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.