Multi-target threshold image segmentation method for fuzzy information and statistical information in fusion interval

Abstract

The invention discloses a multi-target threshold image segmentation method for fuzzy information and statistical information in a fusion interval. The method comprises the steps of: inputting an image to be segmented and converting the image into a grayscale image; setting the initial population number of the image as N, the maximum number of iterations as G and the maximum threshold number as Smax, and then dividing the population into several grouping population Qs of the same size according to the threshold number; conducting multi-target evolution for the obtained grouping population Qs through simultaneous optimization of the interval modulus entropy function and the inter-class variance function based on the linear intercept histogram to allow each grouping population to obtain a group of non-dominated solution sets, selecting an optimal solution in the non-dominated solution set of each grouping population through weighted ratio of the inter-class variances, the optimal solution being the optimal threshold number and the optimal threshold value; and conducting mark assignment for the pixel points in the original image according to the optimal solutions, and obtaining the final segmentation results. The method can realize the adaptive threshold image segmentation, and a satisfactory result can be obtained for a noisy image.

Description

technical field [0001] The invention belongs to the technical field of image processing, and relates to a multi-object threshold image segmentation method for fusing interval fuzzy information and statistical information, in particular to a multi-object evolution adaptive multi-threshold segmentation method based on image interval fuzzy entropy and improved OTSU. Background technique [0002] Image segmentation is one of the most basic and important technologies in image processing and early vision. Image segmentation is the process of dividing an image into several specific regions with unique properties, and extracting the object of interest. Since the 20th century, researchers have proposed many image segmentation methods, mainly including threshold methods, region methods, and clustering methods. Among many image segmentation techniques, the threshold method has become one of the most important and effective techniques because of its simple implementation, small amount ...

AI Technical Summary

Problems solved by technology

Although the above method can suppress the noise in the image to a certain extent, its image segmentation speed and performance are not ideal.

Moreover, the threshold number of the above method is set in advance, and it cannot adapt to the appropriate threshold number as the image changes.

Images