This project is a computer vision-based License Plate Recognition (LPR) system that detects vehicle license plates and extracts text from them. It uses YOLO (Ultralytics) for object detection and EasyOCR for optical character recognition.
The system can process both images and videos to identify license plates in real-time and display the recognized text.
- Real-time license plate detection
- OCR-based text extraction using EasyOCR
- Supports both image and video input
- Fast and efficient detection using YOLO
- Image preprocessing with OpenCV for better accuracy
- Python
- OpenCV
- Ultralytics YOLO
- EasyOCR
LPR-Project/ │── images/ # Input images │── videos/ # Input videos │── output/ # Output results │── models/ # YOLO model files │── main.py # Main execution file │── requirements.txt # Dependencies │── README.md # Project documentation
-
Clone the repository git clone https://github.com/your-username/lpr-project.git cd lpr-project
-
Install dependencies pip install -r requirements.txt
Run on Image: python main.py --image path_to_image.jpg
Run on Video: python main.py --video path_to_video.mp4
- YOLO model detects license plate region
- Detected region is cropped
- Preprocessing applied using OpenCV
- EasyOCR extracts text from plate
- Output displayed with bounding box and text
- Improve OCR accuracy in low-light conditions
- Add support for multiple country plate formats
- Deploy as a web application
- Integrate database for vehicle tracking
This project is open-source and available under the MIT License.