A single-pose series sphere-based calibration method for camera-projector structured light system

Camera-projector structured light sensor plays a crucial part in three-dimensional (3D) reconstruction and robot vision fields. Calibration of the camera-projector structured light system is essential for further measurement. The common-used plane-based calibration method is always not flexible due to the inevitable manual adjustment of multi-pose plane. In this paper, a single-pose series sphere-based calibration method is proposed, which mainly includes two steps. The first step is to calibrate the camera's intrinsic parameters. An improved calibration method based on the image of absolute conic (IAC) is proposed in this step to achieve higher calibration accuracy, and the minimum number of spheres used in this improved method is analyzed. The second step is to calibrate the projector's intrinsic and extrinsic parameters. In order to achieve the 3D coordinates of feature points on the spherical surface, an accurate method based on the back projection theory is proposed in this step. Finally, calibration experiment for a camera-projector structured light system is implemented, and the experimental result demonstrates that the proposed method can achieve desired flexibility, robustness and accuracy.

Automated summary

This paper proposes a camera-projector structured light system calibration method based on a single-pose sphere, which includes two steps: calibration of the camera's intrinsic parameters and calibration of the projector's intrinsic and extrinsic parameters. The experimental results demonstrate that the proposed method achieves the desired flexibility, robustness and accuracy.