Skip to content

Latest commit

 

History

History
60 lines (45 loc) · 1.58 KB

README.md

File metadata and controls

60 lines (45 loc) · 1.58 KB

ANPR-YOLOv8

Automatic number plate recognition (Iranian)

سامانه تشخیص پلاک ایرانی

Results - نتایج

Image 2 Image 3
Image 2 Image 3

Dataset

You can download the dataset from my roboflow profile: https://universe.roboflow.com/barzansaeedpour/anpr-iranian-2

You can also use the code in (download_dataset.py) to download it.

How to use

1- create a virtual environment:

python -m venv env

2- Activate it:

env\Scripts\activate

4- install ultralytics:

pip install ultralytics

5- install pytorch with cuda support:

pip install torch --upgrade torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118

6- Train the model

yolo task=detect mode=train epochs=50 data='dataset/data.yaml' model=yolov8m.pt imgsz=300 batch=2 

7- Test the model

yolo task=detect mode=predict model="runs/detect/train/weights/best.pt" save=True conf=0.3 source='dataset/test/images/'

Truck scale results - Plate Localization (نمونه تشخیص قالب پلاک در باسکول)

Image 2 Image 6

Truck scale results - Character Detection (نمونه تشخیص حروف پلاک در باسکول)

Image 2 Image 3