A Stereo Matching Method Based on Image Segmentation and Adaptive Weight

Abstract

The invention discloses a stereo matching method based on image segmentation and adaptive weight, including disparity initialization and disparity optimization. The cost function of segmentation information and adaptive weight calculates the matching cost E(p,pd,d) of the current pixel point p to be matched in the left image and all candidate matching points pd in the right image, and selects the candidate matching point with the minimum matching cost as p The optimal matching point; Repeat the above steps to traverse each pixel in the left image in raster scanning order to obtain the initial parallax image; parallax optimization includes parallax plane fitting, abnormal suppression and edge repair on the obtained initial parallax image; its advantages It is to re-correct the untrustworthy points in the initial parallax image calculated by the optimal matching point, merge the abnormal small area into the adjacent normal area, and repair the edge pixels to eliminate the parallax error and improve the matching accuracy.

Description

technical field [0001] The invention belongs to the field of image processing, and relates to a stereo matching technology, in particular to a stereo matching method based on image segmentation and adaptive weight. Background technique [0002] In recent years, as one of the hottest research issues in the field of computer vision, stereo vision technology has been widely used in visual navigation, object recognition and industrial control. Stereo vision technology mainly includes image acquisition, camera calibration, feature extraction, stereo matching, and 3D reconstruction. Stereo matching is the core part of stereo vision technology. Correspondence. Whether the images can be accurately matched and the correct three-dimensional coordinates of the scene can be obtained is the key to the success of the stereo vision technology. [0003] According to different constraints, stereo matching can be divided into two categories: global stereo matching and local stereo matching....

AI Technical Summary

Problems solved by technology

For example: a stereo matching method based on the confidence support window, first use the SAD (Sum of Absolute Difference, sum of absolute difference) algorithm to obtain the initial parallax of the pixel, and then use the pixels with higher confidence in each matching window to perform Plane fitting and get the final disparity image, although this method can obtain better matching results, but it is only suitable for image areas with smooth texture, so there are great limitations; a stereo matching algorithm based on joint histogram, The algorithm adopts the method of effectively sampling the pixels in the matching window to reduce the redundant calculation of repeated filtering. Although a fixed spatial sampling value can be obtained, the matching result is too sensitive to the characteristics of the input image and the spatial sampling rate. , this method is not universal at present; a stereo matching method based on an adaptive window, first uses the Gaussian mixture model to describe the disparity distribution of pixels in the matching window, and then determines the size of the matching window according to the uncertainty of the disparity distribution, Although this method improves the matching quality, it also greatly increases the complexity of the algorithm; a matching method based on adaptive support weights, which does not change the size and shape of the matching window, but uses a fixed-size rectangular window, According to the color and distance difference between each pixel point and the center point in the window, the support weight is assigned to carry out energy aggregation. This method effectively avoids the problem of matching window selection. Although it can achieve better matching results, it still has the following shortcomings: For matching In the low-textured areas, structurally repeated areas, and discontinuous areas of disparity in the window, it is difficult to obtain correct matching results by matching based on pixel color and distance

Images