High-precision camera calibration based on concentric circle compensation for industrial environments

Accurate calibration of cameras in industrial production vision systems is a critical fundamental task. However, industrial visual measurement systems often face challenges such as large fields of view, shallow depths of field, and the use of imprecise large calibration templates. These factors make the task of accurate visual measurement in the industrial production environment a great challenge. This paper presents a camera calibration algorithm based on the eccentricity error of concentric circles and the fixed topological relationship constraints of the calibration board structure. In this calibration algorithm, the homography relationship of the calibration board targets is calculated to iteratively optimize the eccentricity error of the concentric circle patterns, providing stable and accurate feature point information for precise camera calibration in industrial settings. Additionally, during the iterative calibration process, deviation parameters are introduced for each feature point on the calibration board relative to the standard plane to account for the machining and geometric deformation errors of the calibration board. This approach addresses issues related to the imprecise calibration of large planar templates. These deviation parameters and eccentricity errors of the concentric circle feature points are optimized together with the camera calibration parameters to correct the positions of the feature points and enhance the camera calibration accuracy in complex industrial scenarios. The results of simulation and experiments validate the feasibility and operability of the proposed camera calibration method. It can fundamentally eliminate perspective transformation errors and improve the precision of camera parameters and target geometry.