Industrial Computer Vision with AI: Defect Detection and Visual Inspectionのトレーニングコース

Introduction to Industrial Computer Vision

• Overview of machine vision systems in manufacturing
• Typical defects: cracks, scratches, misalignments, missing components
• AI vs traditional rule-based visual inspection

Image Acquisition and Preprocessing

• Camera types and image capture settings
• Noise reduction, contrast enhancement, and normalization
• Data augmentation for training robustness

Object Detection and Segmentation Techniques

• Classical approaches (thresholding, edge detection, contours)
• Deep learning methods: CNNs, U-Net, YOLO
• Choosing between detection, classification, and segmentation

Defect Detection Model Development

• Preparing annotated datasets
• Training defect classifiers and segmenters
• Model evaluation: precision, recall, F1-score

Deployment in Industrial Settings

• Hardware considerations: GPUs, edge devices, industrial PCs
• Real-time inspection pipeline architecture
• Integration with PLCs and factory automation systems

Performance Tuning and Maintenance

• Handling changing lighting and production conditions
• Model retraining and continual learning
• Alerting, logging, and QA reporting integration

Case Studies and Domain Applications

• Defect detection in automotive assembly and welding
• Surface inspection in electronics and semiconductors
• Label and packaging verification in pharma and food

Summary and Next Steps