https://github.com/0xnu/uklpr
UKLPR is a computer-vision model architecture purpose-built for detecting, reading, and recognizing United Kingdom license plates.
https://github.com/0xnu/uklpr
computer-vision image-recognition object-detection ocr transport transportation uk united-kingdom
Last synced: 9 months ago
JSON representation
UKLPR is a computer-vision model architecture purpose-built for detecting, reading, and recognizing United Kingdom license plates.
- Host: GitHub
- URL: https://github.com/0xnu/uklpr
- Owner: 0xnu
- License: other
- Created: 2025-08-18T08:25:41.000Z (10 months ago)
- Default Branch: main
- Last Pushed: 2025-08-18T14:22:11.000Z (10 months ago)
- Last Synced: 2025-08-19T03:51:45.341Z (10 months ago)
- Topics: computer-vision, image-recognition, object-detection, ocr, transport, transportation, uk, united-kingdom
- Language: Python
- Homepage: https://0xnu.github.io/uklpr/
- Size: 3.31 MB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
- Codeowners: CODEOWNERS
- Security: SECURITY.md
Awesome Lists containing this project
README
## uklpr
[](/LICENSE)
UKLPR is a computer-vision model architecture purpose-built for detecting, reading, and recognizing United Kingdom license plates. It is optimized for speed and accuracy across diverse UK plate formats.
### Model Performance
- **Detection Rate**: 100.0%
- **Text Extraction Rate**: 100.0%
- **Processing Speed**: 8.1 FPS
- **Model Size**: YOLOv8 Nano (~12.3MB)
### Supported Languages
- English (en)
### Quick Start
#### Installation
```python
pip install ultralytics easyocr opencv-python pillow torch torchvision huggingface_hub
```
#### Usage
```python
import cv2
import numpy as np
from ultralytics import YOLO
import easyocr
from PIL import Image
from huggingface_hub import hf_hub_download
import warnings
# Suppress warnings
warnings.filterwarnings('ignore')
# Download models from HuggingFace
print("Downloading model from HuggingFace...")
model_path = hf_hub_download(repo_id="0xnu/uk-license-plate-recognition", filename="model.onnx")
config_path = hf_hub_download(repo_id="0xnu/uk-license-plate-recognition", filename="config.json")
# Load models with explicit task specification
yolo_model = YOLO(model_path, task='detect')
ocr_reader = easyocr.Reader(['en'], gpu=False, verbose=False)
# Process image
def recognize_license_plate(image_path):
# Load image
image = cv2.imread(image_path)
image_rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
# Detect license plates
results = yolo_model(image_rgb, conf=0.5, verbose=False)
plates = []
for result in results:
boxes = result.boxes
if boxes is not None:
for box in boxes:
# Get coordinates
x1, y1, x2, y2 = box.xyxy[0].cpu().numpy()
# Crop plate
plate_crop = image_rgb[int(y1):int(y2), int(x1):int(x2)]
# Extract text
ocr_results = ocr_reader.readtext(plate_crop)
if ocr_results:
text = ocr_results[0][1]
confidence = float(ocr_results[0][2]) # Convert to native Python float
plates.append({'text': text, 'confidence': confidence})
return plates
# Usage Example
results = recognize_license_plate('sample_car_with_license.jpeg')
print(results)
```
### Model Architecture
#### Detection Model (YOLOv8n)
- **Architecture**: YOLOv8 Nano
- **Parameters**: ~3M
- **Input Size**: 640x640 pixels
- **Output**: Bounding boxes for license plates
#### OCR Model (EasyOCR)
- **Engine**: Deep learning-based OCR
- **Languages**: English
- **Character Set**: Alphanumeric + common symbols
### Training Details
- **Dataset**: UK License Plate Dataset ([0xnu/uk-licence-plate](https://huggingface.co/datasets/0xnu/uk-licence-plate))
- **Training Epochs**: 10
- **Batch Size**: 16
- **Image Size**: 640x640
- **Optimizer**: AdamW
- **Framework**: Ultralytics YOLOv8
### Use Cases
- Traffic monitoring systems
- Automated parking management
- Law enforcement applications
- Toll collection systems
- Vehicle access control
### Limitations
- Optimized for European license plate formats
- Performance may vary with extreme weather conditions
- Requires good image quality for optimal text recognition
- Real-time performance depends on hardware capabilities
### License
This project is licensed under the [Modified MIT License](./LICENSE).
### Citation
If you use this model in your research or product, please cite:
```bibtex
@misc{uklpr2025,
title={UKLPR: United Kingdom License Plate Recognition},
author={Finbarrs Oketunji},
year={2025},
publisher={Hugging Face},
howpublished={\url{https://huggingface.co/0xnu/uk-license-plate-recognition}}
}
```
### Copyright
Copyright (C) 2025 Finbarrs Oketunji. All Rights Reserved.