https://github.com/jirayuwat12/drowsiness-detection
Class project in Computer Vision Class. This project will read the input stream or video and classify that the people is drowsy or not frame by frame.
https://github.com/jirayuwat12/drowsiness-detection
computer-vision opencv python
Last synced: 5 months ago
JSON representation
Class project in Computer Vision Class. This project will read the input stream or video and classify that the people is drowsy or not frame by frame.
- Host: GitHub
- URL: https://github.com/jirayuwat12/drowsiness-detection
- Owner: jirayuwat12
- Created: 2025-02-15T14:20:02.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-03-27T03:02:48.000Z (over 1 year ago)
- Last Synced: 2025-03-27T04:21:52.521Z (over 1 year ago)
- Topics: computer-vision, opencv, python
- Language: Python
- Homepage:
- Size: 3.22 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# drowsiness-detection
## Installation
1. Clone the repository
2. Change the directory to the repository
```bash
cd drowsiness-detection
```
3. Install the modules
```bash
pip install .
```
> or if you need to install the modules in development mode
> ```bash
> pip install -e ".[dev]"
> ```
## Usage
1. Run the script for the video mode
> This script will run the drowsiness detection on the video file
respectively to the configuration file.
```bash
python ./scripts/run_drowsiness_detection.py --mode video\
--config ./path/to/config.yaml\
--log_folder ./path/for/log_folder
```
## Dataset
In this project, we have used 2 datasets:
1. [Drowsiness Detection Dataset, *Prasad V Patil*](https://www.kaggle.com/datasets/prasadvpatil/mrl-dataset/data)
2. [Driver Drowsiness Detection (DDD), *Ismail Nasri*](https://www.kaggle.com/datasets/ismailnasri20/driver-drowsiness-dataset-ddd)
## Reference
1. [MediaPipe by Google](https://mediapipe-studio.webapps.google.com/studio/demo/face_detector)
- We use MP to create our dataset(extract face and eye position from our video)