Improved system calibration for specular surface measurement by using reflections from a plane mirror

Area of Science:

• Optical Metrology
• Computer Vision
• Surface Characterization

Background:

• Accurate calibration is crucial for specular surface metrology.
• Existing methods can be complex and prone to imaging aberrations.
• Phase measuring deflectometry requires precise system calibration for reliable measurements.

Purpose of the Study:

• To introduce a flexible, simple, and robust system calibration method for specular surface metrology.
• To improve the accuracy and efficiency of calibration for phase measuring deflectometry.
• To reduce the impact of imaging aberrations and enhance anti-noise ability.

Main Methods:

• A three-step calibration procedure involving camera calibration with plane patterns.
• Measurement of reflection ray directions using correspondence matching.
• Pose estimation via orthogonal iteration algorithm and plane mirror reflections.
• Utilizing reflection rays to replace camera center coordinates, alleviating imaging aberration.
• Employing global optimization with orthogonal projection for precise initial values in bundle adjustment.

Main Results:

• Simulations show absolute errors within 3 pixels for camera parameters and relative errors below 0.5% for screen pose with 0.6-pixel noise.
• The method demonstrates strong anti-noise ability due to reflection rays and global optimization.
• Experimental reconstruction accuracy improved by 60.11% compared to methods using only bundle adjustment.

Conclusions:

• The proposed calibration method is valid, efficient, and robust for specular surface metrology.
• It offers a low-cost, simple setup, providing an economical and flexible approach for phase measuring deflectometry.
• The technique achieves high accuracy and robustness in practical applications.