引力搜索算法在多阈值图像分割中的应用研究

Research on Application of Gravitational Search Algorithm in Multi Threshold Image Segmentation

针对求解高维阈值图像分割计算复杂的问题,提出了一种基于引力搜索算法的多阈值图像分割方法,该方法以大津法(Otsu)设计为适应度函数,利用引力搜索算法快速搜索得到待分割图像的最优阈值,然后根据最优阈值进行图像分割。结合人眼视觉可知,引力搜索算法能够结合应用于图像分割,且能取得较好的效果。实验测试结果表明,该方法与布谷鸟算法、人工蜂群算法比较,引力搜索算法的收敛速度更快,寻优的阈值质量较高。

In order to solve the complex problem of image segmentation of high dimensional threshold,a multi-threshold image segmentation method based on gravity search algorithm is proposed,the method designs the fitness by Otsu,and uses the favorable performance of gravitational search algorithm to fast and accurately find the optimal threshold of the image to be segmented.Then the image segmentation is carried out according to the optimal threshold.Combining with human vision,the gravitational search algorithm can be used in image segmentation and can achieve better results.Experiment results show that compared with cuckoo search algorithm and artificial bee colony algorithm,the performance of gravitational search algorithm is optimal in terms of its convergence speed and better quality threshold.