Sleep Detector

A computer vision application that detects drowsiness by monitoring eye state using facial recognition.

Overview

This script uses OpenCV to detect faces and track eye state (open/closed) in real-time video, designed to alert drowsy drivers.

Features

• 👁️ Eye State Detection — Monitors if eyes are open or closed
• 😴 Drowsiness Alert — Detects when driver may be falling asleep
• 📹 Real-time Processing — Works with live webcam feed
• 🎯 Face Tracking — Locates and tracks face in frame

Requirements

• Python 3.x
• OpenCV (cv2)
• dlib (for facial landmarks)
• NumPy

Installation

pip install opencv-python dlib numpy

Usage

python sleep_detector.py

How It Works

1. Captures video from webcam
2. Detects face using Haar Cascade / dlib
3. Identifies eye regions
4. Calculates eye aspect ratio (EAR)
5. Alerts if eyes closed for extended period

Applications

• Driver drowsiness detection
• Workplace fatigue monitoring
• Safety systems

License

MIT

---

CI Status

All PRs are checked for:

• ✅ Syntax (Python, JS, TS, YAML, JSON, Dockerfile, Shell)
• ✅ Secrets (No hardcoded credentials)
• ✅ Security (High-severity vulnerabilities)