Improved Squirrel Search Algorithm And Its Application In Image Enhancement

Image enhancement is an essential image processing technique that is fundamental to the application of visual information for a variety of engineering problems.The squirrel search algorithm(SSA)is a swarm intelligent optimization algorithm inspired by the gliding mechanism and dynamic foraging strategy of squirrels.It has the advantages of high search efficiency,strong global search ability and fast convergence speed.Combining SSA with image enhancement methods is beneficial for image enhancement algorithms to find the best solution more quickly in the solution space.Therefore,this paper proposes two kinds of improved SSA and applies them to solve different image enhancement problems.The main work done in this paper is as follows:1)In order to realize the automatic optimization of the optimal parameters for grayscale image enhancement,an adaptive image enhancement method based on an improved squirrel search algorithm is proposed.A bilateral search strategy is introduced into the position updating of squirrels on normal trees to increase the likelihood of obtaining an optimal solution.A cyclone foraging strategy is used to update the position of squirrels on acorn trees to improve the convergence rate and search accuracy of the algorithm.In addition,the proposed squirrel search algorithm based on bilateral searching and cyclone foraging(BCSSA)is compared with bat algorithm(BA),differential evolution(DE)algorithm,flower pollination algorithm(FPA),whale optimization algorithm(WOA),squirrel search algorithm(SSA),and four improved squirrel search algorithms on CEC 2017 test functions.The results indicate that BCSSA has higher stability and faster convergence velocity.The proposed BCSSA is applied to grayscale image enhancement and its performance is compared with classical histogram equalization method and SSA by four evaluation indicators,which demonstrate the superiority of BCSSA.2)In order to obtain images with more perceptual details and less noise,an improved squirrel search algorithm is suggested for low-illumination color image enhancement.Using sinusoidal mapping mechanism and opposition-based learning mechanism generate initial population of squirrels to boost the traversal of initial solutions.An adaptive predator presence probability strategy is adopted to balance global exploration and local exploitation capabilities of the algorithm.A mutation operation is introduced to modify the position update formula of squirrels to raise the search efficiency of the algorithm.The proposed squirrel search algorithm based on adaptive predator presence probability and mutation operation(PMSSA)is studied by comparing with four swarm intelligent algorithms and three improved squirrel search algorithms on CEC 2017,CEC 2019 and CEC 2021 test functions with different characteristics.The results reveal that PMSSA has faster convergence and higher solution accuracy.The PMSSA is applied to low-illumination color image enhancement by fusing the two-dimensional gamma transform and the multi-scale retinex with color restoration(MSRCR)so that it can solve the optimal weighting factor of the weighted fusion.On a low-illumination image test set,comparing PMSSA with adaptive histogram equalization(AHE)method,two-dimensional gamma transform,MSRCR,and SSA.The results display that PMSSA can improve overall image brightness and contrast,reduce the effect of uneven illumination,and make the enhanced image clearer and more natural.