Application Research Of Image Processing Based On Improved Swarm Intelligence Optimization Algorithm

In the field of image processing,image segmentation and image enhancement are the first steps of the image processing process.They lay the foundation for the subsequent deeper image processing and recognition.However,the selection of threshold value in traditional image segmentation is a big problem.The selection time is long and the accuracy is low,resulting in poor image segmentation effect.On the other hand,traditional image enhancement methods are inefficient and have weak anti-interference ability,which will lead to poor enhancement effect in darker areas Due to the over-enhancement of brighter areas and blurred images,this paper considers the introduction of swarm intelligence optimization algorithm and proposes improved sparrow search algorithm and improved artificial bee colony algorithm to improve the efficiency of image segmentation and image enhancement.Firstly,a two-dimensional maximum entropy image threshold segmentation based on the improved sparrow search algorithm is proposed.Aiming at the problem that the sparrow search algorithm will fall into the local optimal solution at the later stage of iteration,Levi flight random number and nonlinear inertia weight factor are introduced to improve the sparrow position update formula to improve the search ability of the algorithm.In order to verify the feasibility and practicability of the improved method,First,the improved sparrow search algorithm and other swarm intelligence optimization algorithms were compared under the benchmark test function,and the experimental results proved that the improved sparrow search algorithm has stronger search ability.Then,the two-dimensional maximum entropy was used as the fitness function to segment images under different conditions,and the peak signal-to-noise ratio and feature similarity were used as the evaluation indicators,and compared with other segmentation methods,The experimental results prove the practicability of the improved sparrow search algorithm.After that,a low-illumination image enhancement based on improved artificial bee colony algorithm is also proposed.In order to solve the problem of slow convergence speed of traditional artificial bee colony algorithm and easy to fall into local optimal solution,firstly,population initialization is carried out through Tent chaotic map to improve population diversity,and then the location update formula of hired bees and following bees is improved by combining moth colony algorithm and animal migration algorithm,A reverse learning strategy based on Lambert illumination model is proposed to improve the convergence ability of the algorithm.Through contrast experiments with other swarm intelligence optimization algorithms under the benchmark test function,the experimental results show that the improved method is feasible.Then,combining with incomplete beta function,entropy,edge information and variance as fitness functions,the low-illumination image is enhanced,and compared with traditional artificial colony algorithm and another improved artificial colony algorithm,From the visual observation of the enhancement effect and the distribution of image gray histogram,the improved artificial bee colony algorithm proposed in this paper has better enhancement effect and more uniform gray distribution.