Physical coordinate positioning method based on binocular vision

Abstract

The invention discloses a physical coordinate positioning method based on binocular vision. The physical coordinate positioning method comprises the following steps: S1) setting a left camera and a right camera, establishing a coordinate system of a camera model, and obtaining a conversion relationship between the coordinate system of the camera model and a world coordinate system; S2) extracting a feature point, and obtaining a pixel coordinate of the feature point in a left image and a right image; and S3) calculating a space coordinate of the feature point in the world coordinate system. The left camera and the right camera are used for simulating two eyes, the coordinate system conversion model is established, then, each image shot by the two cameras is subjected to feature point extraction and pixel coordinate calculation, the pixel coordinate is converted into a theoretical coordinate of the camera model, and finally, the space coordinate of a target point is calculated. Measurement accuracy and efficiency is improved, and binocular coordinate positioning can have a better application prospect in the fields including an eye-in-hand system of an industrial robot, industrial cutting, logistics transportation business, packaging business, optical detection and processing and the like.

Description

technical field [0001] The present invention relates to the field of computer vision, more specifically, relates to a physical coordinate positioning method based on binocular vision. Background technique [0002] Binocular stereo vision is an important branch in the field of machine vision. It directly simulates the way human eyes process scenes, which is reliable and simple. Therefore, the binocular stereo vision device is of great application value in many fields such as pose detection and control of the micro-operating system, robot autonomous navigation and aerial survey, three-dimensional metrology and virtual reality. [0003] With the development of machine vision technology, it is more feasible and more advantageous to use the image information collected by the camera to locate the target. Nowadays, positioning based on monocular stereo vision has been widely used. However, monocular vision obtains two-dimensional data, and a large amount of depth information is l...

AI Technical Summary

Problems solved by technology

In the calibration method based on binocular stereo vision positioning, the traditional method can be applied to any camera model, but the calibration accuracy is high. The calibration process is complex, the extraction accuracy is low, the positioning accuracy is low, the efficiency is not high, and the algorithm has low scalability. for a treatment

Images