Calibration method for the rotating axis in panoramic 3D shape measurement based on a turntable

Abstract

In panoramic three-dimensional (3D) shape measurement based on a turntable, the point cloud registration accuracy is greatly affected by the calibration of the rotating axis. To improve the calibration accuracy of the rotating axis, a calibration method based on an auxiliary camera is proposed. The calibration procedure uses a checkerboard and an auxiliary camera. Firstly, the checkerboard is rotated on the turntable to determine the rotation axis, and is used to calibrate the relative orientation of the auxiliary camera and the measurement camera, so that the two cameras have the same world coordinate system. Then, the origin point of the coordinate of the rotation axis is obtained by the auxiliary camera, and the Z-axis direction is taken as the direction of the rotation axis. Finally, the 3D shape of the multi-view is measured by the fringe projection system, and the rotation registration is completed according to the rotation axis vector. The proposed method does not need to reconstruct the 3D coordinates of multiple points when the rotating axis is calibrated. The experimental results show that the proposed method effectively improves the calibration accuracy of the rotation axis vector, and reduces the complexity of the calibration process.

Tool-free axis calibration method for self-rotating three-dimensional vision systems

“…When measuring the large scene or the inner surface of annular workpieces, a rotating mechanism is commonly employed to splice the captured series of data. [7][8][9][10] However, ensuring precise alignment between the camera's optical center and the rotation axis during mounting proves challenging, thus necessitating calibration to establish the positional correlation between the camera and the rotation axis. [11][12][13] Scholars have dedicated significant efforts to the exploration of calibration techniques for rotary scanning system rotation axis.…”

An SVD-based turntable calibration method with multiple reference planes segmented from point clouds

“…Turntable calibration encompasses various crucial aspects, starting with data acquisition, which entails determining the appropriate equipment and methods for acquiring essential data. In turntable calibration, the primary source of data is often derived from images captured from 2D reference objects, with the checkerboard being the most commonly employed reference object [14][15][16][17]. However, extracting precise 3D information from 2D reference objects poses challenges, necessitating researchers to employ diverse measurement techniques.…”

“…In one approach [17], two 2D cameras and two checkerboards are utilized to calculate the relative orientation of a camera. Another study [15] combines a 3D camera with an auxiliary 2D camera to obtain sample points from a checkerboard. To ensure that both cameras share the same coordinate system, specific camera positioning is required, and the checkerboard must be placed horizontally on the turntable plane.…”

“…The second essential aspect of turntable calibration pertains to the selection of an appropriate reference data type. In existing literature, camera-based sampling from 2D or 3D images is commonly employed to acquire reference points for calibration [12,15,[18][19][20][21]. However, extracting points from images introduces inherent random errors, necessitating careful consideration of sample size to ensure method accuracy.…”

A Three-Dimensional Structured Light Vision System by Using a Combination of Single-Line and Three-Line Lasers

“…Therefore, some methods of calibrating the rotation axis have been proposed. Cai [ 31 ] proposed an auxiliary camera-based calibration method. The experimental results show that the method improves the calibration accuracy of the rotation axis vector.…”