Enhanced Harris Hawks Optimizer And Its Applications

Harris hawks optimizer(HHO)is a new swarm intelligence optimization method which imitates the predatory behavior of Harris hawks in nature.HHO has the characteristics of intuitive structure,easy implementation and good global optimization performance.However,the research of HHO is still in its infancy,and there are still some shortcomings,such as slow convergence speed and weak local search ability.This paper analyzes and improves some shortcomings of Harris hawk optimization algorithm,and puts forward some better performance,namely enhanced Harris hawk optimization algorithm.The purpose is to improve and broaden the theoretical basis and application scope of Harris hawk optimization algorithm,and provide an effective new method for solving complex optimization problems.The main work of this paper is as follows:(1)In order to enhance the global search ability of Harris hawk optimization algorithm,balance its global and local search ability better,and improve its convergence accuracy,a Harris hawk optimization algorithm based on neighborhood barycenter reverse learning(NCOHHO)is proposed by introducing neighborhood barycenter reverse learning strategy Ability.NCOHHO is used to test 23 classical test functions,and compared with the classical heuristic optimization algorithm,the experimental results show that ncohho has better performance in function optimization.(2)The training goal of feedforward neural network is to find a group of optimal weights and deviations to achieve the minimum error of the network.Compared with other meta heuristic algorithms,NCOHHO algorithm has a wide detection range and strong mining ability.(3)This paper proposes a small target detection method based on side suppression in complex background.The principle of side suppression in vision is applied to image preprocessing to realize adaptive preprocessing by background suppression and target enhancement.LI-HHO is applied to the image matching problem.Experimental results show that LI-HHO is more effective and robust than other optimization algorithms in solving image matching problems.