Simulation of a machine vision system for reflective surface defect inspection based on ray tracing

Area of Science:

• Optics and Photonics
• Computer Vision
• Materials Science

Background:

• Machine vision systems are crucial for industrial inspection, particularly for detecting defects on reflective surfaces.
• Accurate simulation of optical systems is essential for optimizing inspection processes and reducing experimental costs.
• Challenges in inspecting reflective surfaces include glare and low contrast, necessitating advanced imaging techniques.

Purpose of the Study:

• To develop a comprehensive ray tracing simulation for a machine vision system designed for reflective surface defect inspection.
• To investigate and optimize image fusion techniques for enhancing defect contrast and detection efficiency.
• To validate the simulation's effectiveness by comparing it with experimental results on car painted surfaces.

Main Methods:

• A virtual reflective plane with simulated defects (scratches, pits) was created.
• Monte Carlo ray tracing was employed to generate realistic images, incorporating camera models with finite apertures and bidirectional reflectance distribution functions (BRDFs).
• A flexible fusion method based on differential images was developed and evaluated for defect enhancement, analyzing parameters like image number, light tube width, and fusion interval.

Main Results:

• The simulation accurately reproduced realistic images of the virtual reflective surface with defects.
• The proposed image fusion method effectively enhanced defect contrast against a uniform background.
• Parameter analysis identified optimal settings for the inspection process, improving detection efficiency.
• Experimental validation on car painted surfaces confirmed the simulation's ability to guide optical system setup and defect detection.

Conclusions:

• The developed ray tracing simulation provides a powerful tool for designing and optimizing machine vision systems for reflective surface inspection.
• The simulation significantly reduces the need for extensive experimental trials, saving time and resources.
• The optimized fusion method enhances the reliability and efficiency of detecting surface defects on challenging reflective materials.