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

Area of Science:

• Computer Vision
• Metrology
• Industrial Automation

Background:

• Accurate camera calibration is crucial for industrial vision systems, but faces challenges like large fields of view and imprecise calibration templates.
• Existing methods struggle with the accuracy demands of industrial visual measurement in complex environments.

Purpose of the Study:

• To develop a robust camera calibration algorithm for industrial settings.
• To address challenges posed by large fields of view, shallow depths of field, and imprecise calibration templates.
• To enhance the precision of camera parameters and target geometry in industrial visual measurement.

Main Methods:

• A novel algorithm utilizing eccentricity error of concentric circles and fixed topological constraints of calibration boards.
• Iterative optimization of eccentricity error by calculating homography relationships of calibration targets.
• Introduction of deviation parameters to correct for machining and geometric deformation errors of calibration boards.

Main Results:

• The algorithm provides stable and accurate feature point information for precise industrial camera calibration.
• It successfully corrects feature point positions by optimizing deviation parameters and eccentricity errors.
• Simulation and experimental results validate the method's feasibility, operability, and improved accuracy.

Conclusions:

• The proposed method fundamentally eliminates perspective transformation errors.
• It significantly enhances camera calibration accuracy in complex industrial scenarios.
• The approach offers a more precise solution for industrial visual measurement and metrology.