Gray scale image threshold segmentation method based on symmetric Gamma divergence

Abstract

The invention discloses a gray scale image threshold segmentation method based on symmetric Gamma divergence, and the method comprises the steps: inputting a to-be-segmented image, and solving a normalized gray scale histogram of the to-be-segmented image; building a symmetric Gamma divergence expression of the image before and after segmentation; solving a gray scale value which enables the gray scale value of the expression to be minimum in a gray scale range of the image; carrying out the threshold segmentation of the image through the gray scale value, and outputting a segmented image. The method improves the quality of image segmentation. The edge contour of the segmented image is precise, and the texture detail is clear. The method improves the universality, and is suitable for an image processing task with high requirements for instantaneity.

Description

technical field [0001] The invention relates to the field of image segmentation in machine vision, specifically a threshold segmentation method based on symmetrical Gamma divergence of grayscale image histogram information to realize rapid and accurate segmentation of industrial images such as nondestructive testing based on machine vision. Background technique [0002] Image segmentation is one of the most basic but also the most difficult and challenging problems in image processing. The purpose of image segmentation is to divide the image into multiple regions that do not overlap with each other, and the internal objects in each region are homogeneous, so as to lay the foundation for the subsequent processing of the image. Due to the influence of many factors in the image imaging process, its complexity also makes the segmentation method not universally applicable to different segmentation tasks. Therefore, researching new methods for specific segmentation tasks in practi...

AI Technical Summary

Problems solved by technology

The minimum cross-entropy thresholding method proposed by Li and Lee is the most famous image threshold segmentation method based on the concept of cross-entropy (that is, information divergence, relative entropy). In addition to this method, other famous thresholding methods related to the concept of cross-entropy There is also the minimum error thresholding method proposed by Kittler and Illingworth. This method is essentially a relative entropy method based on the concept of mean square error of Euclidean distance. The mean square error cannot completely and effectively distinguish the relationship between image pixels. Therefore, there are also deficiencies in the segmentation of images

In addition, Chinese scholar Tang Yingqian and others proposed a minimum Tsallis cross-entropy method based on the uniform distribution based on the Tsallis cross-entropy. However, in the real environment, the pixel distribution of the image does not always obey the uniform distribution, so the segmentation performance of the method It also needs to be improved; based on the chi-square divergence (χ 2 -divergence) method is another image thresholding method based on the concept of divergence proposed by related scholars. This method is very sensitive to the distribution of the histogram, and a good threshold cannot be obtained when the histogram is unevenly distributed.

Images