Logistics cargo volume non-contact measurement method based on binocular stereo vision and point laser range finding

Abstract

The invention provides a logistics cargo volume non-contact measurement method based on a computer binocular stereo vision technology and a point laser range finding technology. According to the method, a point laser range finding supplementary means is introduced; a point laser rangefinder is used to acquire the correspondence among the depth of a measurement point on a cargo object, the distance from an initial point and the point coordinate on an image; the incidence relationship between geometric projections of images shot by left and right cameras is established; the geometric projection correlation area of corresponding feature points in left and right images is positioned and set; image stereo matching of left and right image feature points is carried out in the set geometric projection correlation area of the corresponding feature points; and full width feature point extraction is carried out on the left image, which greatly reduces the range of image matching search and the calculation cost. According to the invention, the problem of the uniqueness of the solution of a spatial point reduction problem is solved; the problems of mis-matching and leakage matching in the image matching process are avoided; and the validity and the accuracy of image stereo matching are greatly improved; and the accuracy and the speed of the measurement of a binocular stereo vision system are improved.

Description

technical field [0001] The invention belongs to the technical field of intelligent non-contact measurement and control, and relates to a non-contact measurement method for the volume of logistics goods based on binocular stereo vision and point laser distance measurement. Background technique [0002] The principle of computer binocular stereo vision technology is to accurately match the characteristic imaging points of two images obtained from two cameras with different viewing angles, and then find the accurate projection pixels corresponding to the spatial object points on the left and right images, so as to use triangulation The principle is used to calculate the coordinates of the object point on the space object, so that the size, attitude and spatial position information of the object can be obtained. In terms of measurement, computer vision system technology has the advantages of non-contact, high precision, and fast speed, and can easily obtain its measurement resul...

AI Technical Summary

Problems solved by technology

[0003] However, the gray-scale window matching method generally used in computer image matching using binocular stereo vision technology has a large error in image matching with discontinuous disparity or low detail; Gray-scale similarity calculation, so the calculation is very large, and the processing speed is very slow; because it is also sensitive to camera lens distortion, it is easily disturbed by light and noise, and the algorithm robustness is relatively poor

However, when the image matching under the constraint of epipolar lines is used, when the gray correlation coefficients of the image points to be matched and several image points on the corresponding image epipolar lines are very close, the matched right image matching points will not be real. The imaging points formed by the spatial object points are mis-matched; due to the grayscale changes caused by illumination and noise, the corresponding feature points on an epipolar line in the image cannot be found on the corresponding epipolar line in the related image. Matching points, thus forming a leaky match; when there are several points to be matched on an epipolar line in the image and the gray correlation coefficients of the same image point on the corresponding epipolar line of the related image are very close, it will also cause a false match.

Images