Skip to content

Latest commit

 

History

History
51 lines (34 loc) · 1.64 KB

README.md

File metadata and controls

51 lines (34 loc) · 1.64 KB

Licence plate blurring using YOLOv8_obb

This is my WiP project for licence plate detecting and blurring in images and videos using YOLOv8 oriented bounding boxes (obb). This project is just a hobby project and is not intended for comercial use.

Environment

The code has been developed on:

  • Ubuntu: 22.04
  • CUDA Version: 11.5
  • Python 3.10.12

Dependencies

You can install the requirements with the command:

pip install -r requirements.txt

Pretrained Weights & Dataset

You can download the weights of the pre-trained model from Google Drive and place into the folder weights.

The dataset used for training is a subset of the CCPD dataset.

Inference

For inference, run following command.

Make sure you have added the proper paths in the options.yaml.

  • weights: the path of the weights.pt file
  • input_path: the path of the folder where the images and/or videos you want to test are
  • output_path: the path where you want to save the output of the model
  • save: if you want to save the output of the model keep it to True, else change it to False.
  • show: if you want to preview the results change it to True, else keep it to False.
python main.py

Example

example.mp4

Note: The output video will be in .avi format. If you want to convert it to .mp4 run the following command on terminal:

ffmpeg -i example.avi example.mp4