โจ Key Features
๐ Real-Time YOLO Detection โ Detects PCB defects instantly during inspection
๐ณ๏ธ Defect Categories โ Identifies missing holes, mouse bites, shorts, and spurs
๐ผ๏ธ Image Preprocessing โ Data augmentation, resizing, normalization for better accuracy
๐ค Transfer Learning โ Fine-tuned YOLO models for PCB-specific datasets
๐ Model Evaluation โ Precision, Recall, mAP, confusion matrix for defect classification
โก High-Speed Inference โ Optimized for real-time industrial AOI workflows
๐ ๏ธ Scalable Integration โ Can be embedded into PCB assembly lines with cameras
๐ Visualization Tools โ Bounding boxes, defect heatmaps, and confidence scores
๐งฐ Tech Stack
Programming: Python ๐
Deep Learning Frameworks: YOLOv5 / YOLOv7 / YOLOv8, PyTorch, TensorFlow
Computer Vision: OpenCV, Albumentations
Data Science Tools: NumPy, Pandas, Matplotlib, Seaborn
Deployment: ONNX Runtime, TensorRT, Flask/FastAPI for API integration
๐ Project Structure ๐ dataset/ # PCB defect dataset (train/val/test) ๐ notebooks/ # Jupyter notebooks for model training & evaluation ๐ models/ # Trained YOLO models & weights ๐ src/ # Scripts for preprocessing, training, detection ๐ results/ # Detection outputs, metrics, graphs ๐ deployment/ # Code for real-time inference & API integration
๐ Getting Started git clone https://github.com/yourusername/Deep-Learning-AOI-for-PCBs-Real-Time-YOLO-Detection-of-Missing-Holes-Mouse-Bites-Shorts-and-Spurs.git cd Deep-Learning-AOI-for-PCBs... pip install -r requirements.txt python detect.py --source test_images/
๐ Use Cases
๐ญ Electronics Manufacturing โ Automated PCB inspection on production lines
๐ง Quality Control โ Reduce human errors in PCB defect identification
๐ Research & Development โ Benchmarking PCB inspection using deep learning
๐ Educational Projects โ Learning deep learning applications in industrial AI
๐ค Contributing
Contributions are welcome! Add datasets, improve detection accuracy, or optimize deployment and submit a PR.
๐ License
MIT License โ Open-source for industrial AI innovation and research.
โญ Support
If this project helps you, please star โญ the repo and share it with the AI + Electronics community.