Image multi-threshold segmentation method based on improved artificial raindrop optimization method

The invention discloses an image multi-threshold segmentation method based on an improved artificial raindrop optimization method. The method consists of the steps of preprocessing an image, setting atarget function, finding an optimal threshold with the improved artificial raindrop optimization method, and carrying out image multi-threshold segmentation. A colorful image is read and is subjectedto graying processing to obtain the grey-scale map of the image. The initial position of the raindrop is distributed in the pixel boundary value range of the grey-scale map, and an Otsu method is used as the fitness function of the method to determine the fitness function value of the initial position of each raindrop. Each raindrop is subjected to the process of raindrop falling, raindrop collision and raindrop gathering to continuously generate a new individual and update a raindrop pool. Through multiple iterations, a globally optimal segmentation threshold is found, and the gray level image is subjected to multi-t