https://github.com/divitmittal/driver-drowsiness-detection

Real-time drowsiness detection on driver's face continuously for signs of fatigue using deep learning methodologies
https://github.com/divitmittal/driver-drowsiness-detection

computer-vision deep-learning drowsiness-detection siamese-neural-network

Last synced: 8 months ago
JSON representation

Real-time drowsiness detection on driver's face continuously for signs of fatigue using deep learning methodologies

• Host: GitHub
• URL: https://github.com/divitmittal/driver-drowsiness-detection
• Owner: DivitMittal
• License: mit
• Created: 2024-12-01T15:53:23.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2025-04-18T01:36:43.000Z (about 1 year ago)
• Last Synced: 2025-04-18T14:57:33.819Z (about 1 year ago)
• Topics: computer-vision, deep-learning, drowsiness-detection, siamese-neural-network
• Language: Jupyter Notebook
• Homepage:
• Size: 1.34 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
Driver-Drowsiness-Detection


## Table of Contents

- [Introduction](#introduction)

- [Features](#features)

- [Technologies Used](#technologies-used)

## Introduction

Driver drowsiness is a major contributor to road accidents. This **Driver Drowsiness Detection** system uses **Siamese Neural Networks** and computer vision techniques to monitor a driver's facial features in real-time and detect signs of fatigue. By analyzing eye states, head pose, and mouth movements (including yawning), the system identifies potential drowsiness and enhances road safety.

## Features

- **Real-Time Drowsiness Detection:** Monitors the driver's face continuously for signs of fatigue.

- **Eye Aspect Ratio (EAR):** Utilizes EAR to determine prolonged eye closures, a common sign of drowsiness.

- **Head Pose Estimation:** Detects abnormal head movements such as nodding or tilting.

- **Mouth and Yawning Detection:** Analyzes mouth movements using computer vision techniques to identify yawning, another common fatigue indicator.

- **Siamese Neural Networks:** All detection (eye, head pose, and yawning) is powered by a Siamese Neural Network, which allows the system to effectively compare facial features and detect signs of drowsiness.

## Technologies Used

- **Python 3.8+**

- **Tensorflow & Keras (Siamese Neural Networks)** for feature comparison and detection.

- **OpenCV** for real-time computer vision tasks.

- **Dlib** for facial landmark detection.

- **Imutils** for image processing utilities.