aiInternal Project
Anomaly Product Detection
YOLO-based computer vision model for classifying and detecting anomaly wood products in production with 96% accuracy.
Developed at ProManage
Lead Data Scientist
2019-2020
Internal project - no public access
Screenshots
About This Project
Built a high-accuracy predictive model for classifying and detecting anomalous wood products in a production environment. The system uses YOLO object detection combined with PyTorch and OpenCV to identify defective products in real-time on the production line, enabling immediate quality control actions.
Key Features
- YOLO-based real-time anomaly detection
- Multi-class wood product defect classification
- Production line camera integration
- Automated defect logging and reporting
- Configurable detection sensitivity thresholds
- Visual overlay for detected anomalies
Challenges & Solutions
- Training YOLO on domain-specific wood defect data
- Achieving real-time inference on production hardware
- Handling subtle visual differences between normal and defective products
- Minimizing false positives in high-throughput production
Results & Impact
- 96% accuracy in anomaly product detection
- Real-time quality control on the production line
- Significant reduction in defective product shipments
Technologies Used
PythonPyTorchOpenCVYOLODeep LearningDocker
Project Details
- Category
- ai
- My Role
- Lead Data Scientist
- Duration
- 5 months
- Year
- 2019-2020
- Company
- ProManage
- Status
- Internal / Private
Internal Project
This project was developed for internal use at ProManage. Source code and live demo are not publicly available due to confidentiality.