A method and device for image segmentation based on sparse principal component analysis

Abstract

The invention discloses a method and device for image segmentation based on sparse principal component analysis. For each point in the image to be processed, the neighborhood of the point is obtained, and similar image blocks are obtained according to the gray value of the neighborhood to form a sample. In the training set, the sparse principal component analysis is performed on the sample block corresponding to the point in the sample training set, and the sparse principal component expression base and sparse principal component expression coefficient are obtained. According to the sparse principal component expression base and sparse principal component expression coefficient, the The pixel value of the image block centered on the point does not contain noise, and the noise-free image is constructed according to the noise-free pixel value corresponding to each point in the image to be processed, and the two-dimensional histogram is performed on the noise-free image respectively. Global threshold segmentation and local threshold segmentation based on the moving average method. The images obtained by the two segmentation methods are obtained according to the regional connectivity, and the segmented image is obtained to ensure that the segmented image is not affected by noise.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a method and device for image segmentation based on sparse principal component analysis. Background technique [0002] Image segmentation is the process of dividing an image into regions with specific properties and extracting objects of interest. The images after image segmentation are non-overlapping, which realizes the separation of the target and the background in the image, which is beneficial to the subsequent feature extraction and target analysis of the image. At present, image segmentation technology has been widely used in biomedical images, remote sensing images, and military fields. Especially in machine vision, image segmentation has become a research hotspot. Machine vision divides image processing into three levels: bottom layer, middle layer, and high layer. Image segmentation links the bottom layer feature processing and high-level processing o...

AI Technical Summary

Problems solved by technology

[0003] In the prior art, due to the noise in the image, the image segmentation is affected by the noise, resulting in the degradation of the image segmentation quality, affecting the visual effect of the image, and even covering up some features of the image, which directly affects the subsequent processing effect of the image.

Images