Color image three-dimensional reconstruction method based on three-dimensional matching

Abstract

The invention relates to a color image three-dimensional reconstruction method based on three-dimensional matching, comprising the following steps of: (1) simultaneously and respectively taking an image from proper angles by using two color cameras; (2) respectively calibrating the internal parameter matrixes and the external parameter matrixes of the two cameras; (3) carrying out polar line correction and image transformation according to calibrated data; (4) working out matching cost for each pixel point in the two corrected images by applying a self-adaption weight window algorithm and acquiring an initial parallax image; (5) marking the reliability coefficient of the pixel initial matching result by adopting matching cost reliability detection and left and right consistency verification; (6) carrying out color segmentation on the images through a Mean-Shift algorithm; (7) carrying out global optimization by a selective confidence propagation algorithm on the basis of color segmentation and pixel reliability classification results to obtain a final parallax image; and (8) working out the three-dimensional coordinates of actual object points on the images according to the calibrated data and the matching relation, thereby reconstructing the three-dimensional point cloud of an object.

Description

technical field [0001] The invention belongs to the technical field of binocular stereo vision, relates to the problem of processing color real-shot images based on stereo matching, and in particular relates to a method of using an adaptive weight algorithm to calculate matching costs and selectivity based on image segmentation and pixel reliability classification results The belief propagation algorithm is used to obtain the matching relationship of pixel points to reconstruct the three-dimensional point cloud of the target object. Background technique [0002] Binocular stereo vision technology is a passive three-dimensional measurement method, which is flexible in implementation, low in environmental requirements, and friendly to human-computer interaction. It is a relatively popular technology in three-dimensional reconstruction algorithms. Binocular stereo vision aims to simulate the mechanism of human binocular recognition of three-dimensional information of the scene,...

AI Technical Summary

Problems solved by technology

[0007] In the one-dimensional scanning line, the smoothness constraints in other directions are lost. The graph cut algorithm takes a long time, and it is difficult to meet the real-time requirements of 3D reconstruction of real-shot images. The belief propagation algorithm indiscriminately Confidence is propagated between disparity discontinuities, and disparity continuity constraints may not be satisfied between neighboring pixels in the disparity discontinuity area, resulting in unclear borders of the reconstructed point cloud

[0008] Due to the above shortcomings, the existing 3D reconstruction algorithms based on stereo matching cannot obtain satisfactory results in practical applications.

Images