Study On Image Segmentation Algorithm Based On Improved Swarm Intelligence Optimization Algorithm

Image segmentation can essentially be regarded as the problem of finding the best parameter value to segment the image in a complex data space.The swarm intelligence optimization algorithm has unique advantages in solving complex nonlinear multi-modal function optimization problems.This paper mainly studies to improve the efficiency of image thresholding segmentation and optimization.An improved krill herd optimization algorithm and an improved constrained chicken swarm optimization algorithm are proposed and applied to image threshold segmentation.The specific research work is as follows:First,an improved krill herd algorithm based on time-varying nonlinear descending strategy is proposed to avoid the difficulty of the basic krill herd algorithm being trapped in the local pole value,and the global search ability of the algorithm is improved.Simulations were performed with standard test functions and compared with the others algorithm.The results proved the rationality and effectiveness of the proposed improved krill herd optimization algorithm.After that,the standard image is used as the image segmentation object,and the maximum class variance is the fitness function.The improved krill herdoptimization algorithm is applied to the image threshold segmentation.The comparison with the others algorithm shows that the proposed improved krill group optimization algorithm is more efficient for image threshold segmentation.Second,an improved constrained chicken swarm optimization algorithm was proposed,which improve the optimization performance and convergence of the basic chicken swarm optimization algorithm from the two aspects of constraint function and evolution mechanism.Compared with the others algorithm,the improved chicken group optimization algorithm is proved to be superior.Then the improved constrained chicken group optimization algorithm and the maximum class variance combination should be used for image threshold segmentation.Compared with the others algorithm,the proposed improved chicken group optimization algorithm can well improve the efficiency of image threshold segmentation optimization.