Binocular stereoscopic vision image measurement method based on polar line correction

Abstract

The invention provides a binocular stereoscopic vision image measurement method based on polar line correction, used for solving a technical problem of low measurement precision of a binocular stereoscopic vision image in the prior art. The implementation steps of the method comprises: (1) acquiring internal parameters and external parameters of a binocular camera; (2) performing polar line correction on the binocular stereoscopic vision image; (3) preprocessing the binocular stereoscopic vision image subjected to polar line correction; (4) obtaining a matching point pair set of the preprocessed binocular stereoscopic vision image; (5) calculating the three-dimensional coordinates of all points to be measured in the target object corresponding to the matching point pair set under the worldcoordinate system; and (6) obtaining geometric distance information between any two points of the target object in the binocular stereoscopic vision image. According to the method, a polar line correction algorithm is added in the binocular stereoscopic vision image measurement process, the binocular stereoscopic vision image measurement precision is improved, and the method can be used for sizemeasurement of a target object with the high precision requirement.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and relates to a binocular stereo vision image measurement method, in particular to a binocular stereo vision image measurement method based on epipolar correction, which can be used for dimension measurement of target objects requiring high precision. [0002] technical background [0003] In actual production and life, people have a great demand for the size measurement of objects, and the requirements for measurement efficiency and measurement accuracy are getting higher and higher. The traditional measurement of object size mostly adopts the manual contact measurement method. The actual measurement is carried out with the help of measuring tools such as a tape measure, and the measurement data is manually interpreted and recorded. For high target objects, this method is not easy to operate and has certain risks. [0004] In order to solve the disadvantages of low efficiency, poor oper...

AI Technical Summary

Problems solved by technology

This method does not need a reference object with known size in the scene, and effectively overcomes the limitation of the reference object in monocular vision image measurement, but the disadvantage is that in the process of calculating the disparity value, due to the two images are not realized Line alignment, resulting in a certain error in the parallax value, affecting the measurement accuracy

Images