vehicle number plate detection and ocr using YOLO .
The identifier's approach is straightforward:
- Determine if a number plate is in the picture
- Identify where that number plate is
- Crop out the relevant asset
- Read characters in the cropped picture
- LabelReader uses the Yolov3 algorithm for object detection.
- Darknet (Fast, C Implementation)
- A pretrained model can be downloaded
- Link : "https://drive.google.com/open?id=1-48BHxaXTEZv3nwZEN7jAbdPLi2x40v7".
- Download the model and copy it to the directory
- For custom training follows this tutorial "https://medium.com/@manivannan_data/how-to-train-yolov2-to-detect-custom-objects-9010df784f36"
-
If you are already installed Anaconda python you can skip this step https://www.anaconda.com/distribution/
-
conda create -n yolo pip python=3.7
-
conda activate yolo
-
pip install -r requirements.txt
-
apt-get install tesseract-ocr
-
./darknet
Run the main script
- python app.py