Stereo matching method by utilizing graph theory-based image segmentation algorithm

Abstract

The invention discloses a stereo matching method by utilizing a graph theory-based image segmentation algorithm. The method has the advantages of greatly improving real-time property and ensuring higher robustness on illumination noise compared with a mean-shift segmentation-based image segmentation algorithm. The method comprises the following steps of: 1) acquiring left and right images of an object to be matched and calibrating the images; 2) respectively calculating an initial parallax error of each pixel point in the calibrated left and right images by using a window method; 3) comparingcorresponding parallax error values in the two obtained parallax error images, and selecting a preferable parallax error of each pixel as an initial parallax error in a step 5); 4) segmenting the calibrated left and right images in the step 1) by utilizing the graph theory-based image segmentation algorithm; 5) performing median filtering on initial parallax error image information acquired in the step 3) by utilizing the segmented images; and 6) obtaining a parallax error image of the left and right images to be matched.

Description

technical field [0001] The invention relates to a stereo matching method using an image segmentation algorithm based on graph theory. Background technique [0002] Stereo matching is a core component in the field of computer vision, and it is widely used in the practice of automobile assisted driving and 3D TV. According to the different matching primitives, it can be roughly divided into feature-based matching algorithm and area-based matching algorithm. Although the feature-based stereo matching method is faster, it cannot obtain the global optimal disparity, so the global effect is better in recent years. Good area-based matching algorithms are more widely used, among which graph-cuts, Dynamic Programming and Belief propagation algorithms greatly improve the precision and accuracy of stereo matching. [0003] Tao H, Sawhney H S and Kumar R published the paper "A A Global Matching Framework for Stereo Computation Based on Color Image Segmentation" (A Global Matching Fram...

AI Technical Summary

Problems solved by technology

[0004] In recent years, segmentation-based matching algorithms generally use the mean-shift segmentation algorithm to obtain the segmentation information of the reference image. The mean-shift algorithm is a non-parametric estimation method that uses the gradient of the probability distribution to find the peak value. High robustness, it divides the image by clustering the color domain and spatial domain features of the input color image, the algorithm can get a better segmentation effect, but the real-time performance of the algorithm is general

Images