馃摝 Initial Release
We are proud to present the first official release of the Parcel Damage Detection System. This computer vision solution is designed to automate quality control in logistics by identifying damaged packages in real-time.
馃攳 Key Capabilities
- Automated Inspection: Instantly detects damage on parcels from images or video streams.
- Multi-Class Detection: Identifies specific types of damage such as Dents, Scratches, and Crushed corners.
- High Precision: Powered by a custom-trained YOLO model for accurate bounding box regression.
- Visual Feedback: Draws clear bounding boxes and confidence scores around detected defects.
馃洜 Tech Stack
- Core: Python 3.x
- Model: YOLO (You Only Look Once) architecture
- Vision Library: OpenCV
- Training Framework: PyTorch
馃殌 Getting Started
- Clone the repository.
- Install dependencies:
pip install -r requirements.txt
- Run the detection script:
python detect.py --source path/to/image.jpg
Created by Wiktor Goszczy艅ski