run_20250113_025902.mp4

Pothole Detection

This project uses a YOLOv8 model to detect potholes in images and videos.

Project Structure

Pothole-detection/
├── artifacts/             # Sample videos and images to test
├── results/               # Directory to save the output images and videos
├── weights/               # Contains yolo model weights
├── main.py                # Main script for running the detection
├── helper.py              # helper functions
├── experiments.ipynb      # Jupyter notebook contains step by step training implementation and result metrics
└── requirements.txt       # required packages

Usage

1. Clone repo

git clone https://github.com/amaanrzv39/Pothole-detection.git
cd Pothole-detection

2. Setup virtual env

python -m venv venv
source venv/bin/activate  # On Windows use `venv\Scripts\activate`

3. Install required packages

pip install -r requirements.txt

4. Run

python main.py --source <path_to_image_or_video> --conf <confidence_threshold>

Results on test images

Results on sample videos

https://youtube.com/shorts/2Gc_kqZvPn4?feature=share

https://www.youtube.com/watch?v=ImvIXTIo9I4

https://www.youtube.com/watch?v=9pNrrWmitno

Contributing

Contributions are welcome! If you have ideas or encounter issues, feel free to open a pull request or an issue.

License

This project is licensed under the MIT License.