Image Clustering of Immune Genetic Algorithm Based on Simulated Annealing
This paper proposes an immune genetic algorithm based on simulated annealing for image clustering.Grid template is used to extract the image features of cluster samples,and the solution of the problem is coded.In order to approach the optimal solution,calculate the distance between different antibodies,and construct the vaccine table according to the shortest principle.First,the crossover operator is performed on the population.According to the simulated annealing principle,the variation operator is used to search in the local area.At the same time,in order to accelerate the search speed of the optimized solution,the operation of vaccination operators is carried out.If the new antibody is superior to the old one,it receives; otherwise,it will be received by the Metropolis criterion.The antibody concentration was calculated and the immune balance operator was used to suppress the high concentration of antibodies.According to the fitness and concentration of the antibody,the selection probability was determined.The operation of immune selection operator was carried out,and the combination became a new generation population.The simulation results show that the algorithm can improve the accuracy and efficiency of image clustering.
immune genetic algorithm cluster analysis simulated annealing
Fan Kaixiang Yang Shuying Zhang Zeyi
Tianjin University of Technology,391 Binshui West Road,Xiqing District,Tianjin,China
国际会议
西安
英文
174-181
2018-09-21（万方平台首次上网日期，不代表论文的发表时间）