Research On Image Segmentation Algorithm Based On Wolf Pack Algorithm Optimized By Fractional-order

The traditional wolf pack algorithm(WPA)is easy to fall into local optima and slow convergence in the image segmentation.The Otsu image threshold segmentation algorithm based on fractional order wolf pack algorithm is proposed.Using the gray-gradient two-dimensional histogram,and inter-class variance of Otsu algorithm is defined as the fitness function of the improved WPA to find the optimal segmentation threshold.Using the advantage of fractional order which has memory for past states,the position updating of wolves is controlled by fractional order.The adaptive fractional order is introduced.The position information of wolves is used to adjust the fractional order adaptive,the wandering behavior of the WPA is more reasonable which improves the optimal solution search ability of the WPA and improve the convergence of the algorithm.The particle symmetry distribution method is used to improve the hunting behavior,improve the spatial distribution of the wolves,adjust the position of the wolves during the hunting,and overcome the shortcomings of the local optima in the later stage of the WPA.The target is segmented from the image.The main contents of this paper are:Firstly,research on the advantages and disadvantages of the used image segmentation algorithms,on the principle of the maximum inter-class variance method.The traditional two-dimensional Otsu algorithm can reflect more image information than the one-dimensional Otsu,and the segmentation result is more accurate,but the calculation amount is increased.Therefore,the gray-gradient-based two-dimensional Otsu algorithm is used to separate the noise points of the image from the target background of the image,which not only reduces the search space,but also reduces the amount of computation,and also reduces the noise segmentation results.Impact.Summarize the corresponding evaluation indicators of the image segmentation results.the corresponding evaluation indicators of image segmentation results are summarized.Secondly,the main principles and implementation of the traditional WPA are researched.The influence of the parameters of the WPA on the function of the algorithm and the value range are introduced.The wolf group algorithm has strong robustness,excellent optimal solution,simple parameter setting and population refresh.This paper analyzes the reasons why the traditional WPA is affected by the step value in the wandering behavior and its tendency to fall into the local optima in the later stage of the WPA..Thirdly,the paper studies the superiority of fractional calculus theory in practical application,puts forward the fractional order WPA,the adaptive fractional-order algorithm is introduced into the wandering behavior of the wolf-group algorithm.The positional information of the wolf group is adjusted to the fractional order size,and then the fractional order is used to adjust the position update of the wolf-group wandering behavior.In the siege behavior of the algorithm,the particle symmetric distribution method is introduced to balance the number of wolves at the ends of the prey,which overcomes the shortcomings of the wolves in the siege.It speeds up the convergence of the algorithm and overcomes the disadvantages of the algorithm being easy to fall into the local optimum,applies it on image segmentation.The experimental results show that the improved algorithm proposed in this paper is superior to the traditional image segmentation algorithm.In the number of convergence times,the convergence times of the proposed algorithm are greatly improved compared with the Otsu image segmentation algorithm optimized by the traditional wolf pack algorithm and the Otsu image segmentation algorithm based on the fractional particle swarm optimization.The convergence speed of the algorithm is improved while ensuring the accuracy of image segmentation.