Self-recalibration of a robot-assisted structured-light-based measurement system

Area of Science:

• Computer Vision
• Robotics
• Metrology

Background:

• Structured-light measurement is vital for industrial inspection but requires system movement and reconfiguration.
• Occlusion and object size variations necessitate frequent self-recalibration of extrinsic parameters for accurate 3D scanning.

Purpose of the Study:

• To develop an automatic self-recalibration and 3D reconstruction method for structured-light measurement systems.
• To enable complete workpiece scanning and overcome occlusion issues in industrial inspection.

Main Methods:

• Employed a robot arm to move the structured-light measurement system across different viewpoints.
• Utilized fundamental matrix calculations and point cloud registration for automatic self-recalibration of extrinsic parameters.
• Eliminated the need for an accurate calibration gauge.

Main Results:

• Demonstrated the feasibility of the proposed automatic self-recalibration method.
• Achieved accurate 3D reconstruction of workpieces from multiple viewpoints.
• Validated the system's performance in overcoming occlusion challenges.

Conclusions:

• The proposed method offers an efficient and accurate solution for self-recalibrating structured-light systems in industrial inspection.
• Automated self-recalibration using robot arms and fundamental matrix calculations simplifies the 3D scanning process.
• This approach enhances the adaptability and precision of structured-light measurement for complex industrial applications.