A Matrix Coding Genetic Algorithm Based on Memristor for Image Edge Detection

Xuefeng Zhong, Yongbin Yu, Chen Zhou, Xiangxiang Wang, Xiao Feng, Zhexian Zhou, Jiarun Shen, Jingya Wang, Xinyi Han

Abstract: Researchers have shown the feasibility of memristive networks for some swarm intelligence algorithms like Ant Colony Algorithm (ACO). In this paper, a matrix-based Genetic Algorithm (MGA) for image edge detection is proposed, which means the evolution process of Genetic Algorithm (GA) is transformed to matrix computation in the application of image edge detection. Secondly, memristor circuits are designed and mapping relationship between memristor and GA is built up for the deployment of GA operators. Thirdly, the simulation results of memristor circuits show the feasibility of deploying GA with memristors and the experiment results compared with other memristor-based swarm intelligence algorithms illustrate the effectiveness and superiority of the proposed method. (MCGABM) Finally, two methods are adopted as novel evaluation indicators for the image edge detection to discuss the performance of the proposed algorithm more objective.