Method for calibrating relation between rotating shaft and camera based on constraint global optimization

Abstract

The invention discloses a method for calibrating the relation between a rotating shaft and a camera based on constraint global optimization, and the method comprises the following steps: 1), calibrating internal parameters of a big-view-field camera and a to-be-calibrated camera; 2), taking a series of images of a calibration plate in a 360-degree rotation process of a rotating platform through employing the big-view-field camera; 3), obtaining a rotation matrix and transition matrix of the calibration plate to the coordinate system of the big-view-field camera and a rotation matrix and transition matrix of the calibration plate to the coordinate system of the camera through the images, obtained at step 2), of the calibration plate; 4), obtaining a rotation matrix and transition matrix of a rotary table calibration plate rotating at one place to the coordinate system of the big-view-field camera through the image, taken by the big-view-field camera at step 2), of the rotary table calibration plate, extracting the three-dimensional coordinates of at least two corners of the rotary table calibration plate in the series of images, calculating the circle center coordinates of the track of each corner in a rotating process, carrying out the fitting of the straight line where all circle centers are located, and obtaining the inclined angle between the rotating shaft and the optical axis of the camera according to a rotating shaft direction vector and a camera optical axis direction axis.

Description

technical field [0001] The invention relates to the fields of machine vision, image processing and depth measurement, in particular to a calibration method for the relationship between a rotating shaft and a camera based on constrained global optimization. Background technique [0002] With the advent of the information age, the use of color digital cameras and depth cameras to simultaneously collect the color and depth information of the scene has been more and more widely used in various industries. But this type of camera can only collect data information in one direction. In order to obtain information of a larger range or 360-degree scene, it is necessary to move or rotate the camera to collect a series of two-dimensional or depth data. Especially indoors, the data of a larger field of view is often obtained by rotating the camera around the rotating platform at a known angle. By transforming the data obtained from different viewing angles into the same coordinate sys...

AI Technical Summary

Problems solved by technology

However, when the camera is fixed on the rotating shaft, the camera cannot directly collect images of the calibration plate fixed on the rotating shaft from different angles

Images