Rapid Global Calibration Technology for Hybrid Visual Inspection System.

Area of Science:

• Robotics and Automation
• Computer Vision
• Industrial Metrology

Background:

• Vision-based systems are crucial for efficient industrial product quality inspection.
• Hybrid visual inspection systems, integrating robots and sensors, are vital in mass production, particularly automotive manufacturing.

Purpose of the Study:

• To propose a rapid global calibration method for hybrid visual inspection systems.
• To enhance the accuracy and reliability of industrial inspection processes.

Main Methods:

• Global calibration of a flexible sensor using robot kinematics.
• Global calibration of stationary sensors based on homography, utilizing the calibrated flexible sensor.
• Error compensation for the hybrid inspection system.

Main Results:

• Achieved high accuracy with over 95% of featured points having deviations less than ±0.3 mm.
• Demonstrated strong data correlation with over 85% of points showing correlation coefficients greater than 0.7.
• The calibration process is efficient and easily repeatable for system maintenance.

Conclusions:

• The proposed rapid global calibration method is effective for hybrid visual inspection systems.
• The method significantly improves the accuracy of industrial quality inspection.
• The system's performance is validated against Coordinate Measuring Machine (CMM) standards.