Image threshold segmentation method based on two-dimensional empirical mode decomposition and genetic algorithm

Abstract

The invention relates to an image threshold segmentation method based on two-dimensional empirical mode decomposition and a genetic algorithm. A to-be-segmented image is inputted and the background ofthe image is removed by using a morphological difference method to obtain difference image; decomposition is carried out based on a two-dimensional empirical mode decomposition algorithm to obtain aseries of intrinsic mode functions and remainders; low-frequency sub-images that cannot express detailed information are removed and high-frequency sub-images are screened out; on the basis of a contrast stretching and transformation algorithm, contrast stretching and transformation are carried out on the high-frequency sub-images; the transformed high-frequency sub-images are added and fused to obtain an enhanced image; and threshold segmentation is carried out on the enhanced image by using a genetic algorithm. According to the invention, in terms of segmentation effect, the image thresholdsegmentation method has the improved target segmentation effect and good segmentation integrity by being compared with the way of carrying out genetic-algorithm-based threshold segmentation directly on the image without any distinguishing; and in terms of the feature similarity, the image threshold segmentation method obtains a value higher than that obtained by genetic-algorithm-based threshold segmentation directly on the image.

Description

technical field [0001] The invention relates to an image threshold segmentation method based on two-dimensional empirical mode decomposition and genetic algorithm, belonging to the technical field of digital image processing. Background technique [0002] In recent years, many image segmentation techniques have been proposed, but each has its own advantages and disadvantages. At present, image segmentation technologies at home and abroad mainly include threshold segmentation, region growth segmentation, and edge detection segmentation. Threshold segmentation is a very simple image segmentation method, but it will encounter optimization problems when processing images based on this method, and it is not easy to determine the optimal segmentation threshold; region growth segmentation was first proposed by Levine et al. The image segmentation method of serial area segmentation, although its idea is relatively simple, it can usually segment connected areas with the same charact...

AI Technical Summary

Problems solved by technology

Threshold segmentation is a very simple image segmentation method, but it will encounter optimization problems when processing images based on this method, and it is not easy to determine the optimal segmentation threshold; region growth segmentation was first proposed by Levine et al. The image segmentation method of serial area segmentation, although its idea is relatively simple, it can usually segment connected areas with the same characteristics, and can provide good boundary information and segmentation results, but noise and grayscale inhomogeneity may cause Hole and over-segmentation; Segmentation based on edge detection is to highlight details by seeking first-order or second-order differentiation. Although it can accurately locate the target for segmentation, this method is sensitive to noise

Therefore, although the above method can segment the region of interest in the image, there will be some disadvantages, which will bring trouble to the work in the fields of image analysis and recognition.

Images