Blanking feature constraining calibrating method and device for binocular vision sensor

Abstract

The invention discloses a blanking feature constraining calibrating method and device for a binocular vision sensor. The method comprises the steps of 1, fixing the positions of a left camera and a right camera according to a measuring object, and calibrating internal parameters of the left camera and the right camera; 2, setting a parallel line target, and shooting the target images from different angles by the left and right cameras; 3, correcting the deformation of the images according to the camera deviation coefficients calibrated in step 1; 4, calculating a blanking line according to the linear equation extracted from the target image, and performing the linear calculation method to solve the initial value of the structural parameters of the binocular vision sensor; 5, treating the linearly solved R and T in step 4 as the initial values, and integrally optimizing the consistency under the coordinate systems of the left and right cameras according to the parallel line spacing constraining and the target plane normal direction so as to obtain the final calibration result. With the adoption of the method, the stepped solving of the rotating matrix parameter R and the translation vector parameter T can be achieved, and the difficulty and complexity at calculation can be reduced.

Description

technical field [0001] The invention relates to computer vision technology, in particular to a method and device for calibrating the blanking feature constraint of a binocular vision sensor. Background technique [0002] The binocular vision sensor is mainly composed of two cameras, based on optical triangulation and the principle of stereo parallax, to complete the three-dimensional measurement of feature points and feature lines in the public field of view. Because binocular vision measurement has the advantages of non-contact, fast speed, good system flexibility, and high measurement accuracy, it is widely used in the fields of 3D model reconstruction, 3D information measurement of object surface contours, and measurement of key geometric parameters of objects. [0003] The calibration of the measurement model parameters of the binocular vision sensor is the key to the success of the application of the binocular vision sensor, mainly including the calibration of the inter...

AI Technical Summary

Problems solved by technology

Due to the interaction of different planes on the illumination, the three-dimensional target method can only obtain high-quality calibration images at specific positions, and the three-dimensional target processing is difficult and expensive, so the application of this method in engineering is limited

Based on the unknown moving plane circular hole target method, the calculation process is an iterative process for solving the roots of nonlinear equations, with a large amount of calculation and complicated calculation process

Based on the one-dimensional target method with unknown motion, although the accuracy is high and the implementation is convenient, the solution process requires multiple matrix transformations and iterative root-finding of nonlinear equations, resulting in high computational complexity and large computational errors

In the camera self-calibration method based on feature point matching, the high precision of calibration needs to be based on the high precision and high matching rate of image feature point extraction, which is difficult to guarantee in a complex field environment

In addition, there are some other calibration methods, such as the binocular vision sensor calibration method based on BP neural network proposed by Li Mingzhe in the article "3D measurement system for sheet metal surface based on computer vision", but the calibration accuracy is not given; A linear self-calibration algorithm based on binocular active vision" proposed a self-calibration method based on binocular active vision, but requires a pure translation relationship between the two cameras

Images