🚗 Driver Drowsiness System & Vehicle Counter

A high-performance, real-time computer vision project designed to enhance road safety through Driver Drowsiness Detection and Vehicle Traffic Analysis.

🌟 Key Features

• Real-time Drowsiness Detection: Monitors Eye Aspect Ratio (EAR) and triggers an alarm if fatigue is detected.
• Aural Alerts: Immediate beep sound notification when eyes stay closed for too long.
• Vehicle Counter: Robust traffic analysis counting vehicles crossing a detection line.
• User-Friendly Interface: Real-time feedback with intuitive UI overlays and easy exit options.

---

🛠️ Installation

1. Prerequisites

• Python 3.8 or higher
• A working webcam (for drowsiness detection)

2. Setup

# Install dependencies
pip install -r requirements.txt

Tip

Windows Users: If you face issues installing dlib, we have pre-configured the project to use dlib-bin, which installs instantly without requiring complex C++ build tools.

---

🚀 How to Run & Control

1. Driver Drowsiness System

Monitor alertness in real-time using your webcam.

python dds.py

• How it works: If your eyes stay closed for ~1 second, a "DROWSY ALERT!" appears and a beep sounds.

2. Vehicle Counter

Analyze traffic from the included video file.

python vehicle.py

• How it works: Increments the counter every time a vehicle crosses the horizontal detection line.

---

🚪 How to Close & Shutdown

Each application window can be closed using any of the following methods:

1. Keyboard Shortcuts:
   • Press q to quit immediately.
   • Press ESC to exit.
   • Press Enter (specifically in the Vehicle Counter).
2. Mouse Control:
   • Click the Red "X" Button on the top-right of the window.
3. Terminal Shutdown:
   • If you need to force-stop the script from the command line, press Ctrl + C in your terminal.

---

🧠 Technical Details (EAR Logic)

The drowsiness system calculates the Eye Aspect Ratio (EAR) using 6 facial landmarks per eye.

• Threshold: 0.25
• Consecutive Frames: 20 (triggers alarm if eyes are closed for 20 frames).

---

📁 Project Structure

• dds.py: Main Drowsiness Detection script.
• vehicle.py: Traffic Analysis & Vehicle Counting script.
• face_detection_demo.py: Simple Haar Cascade face detection test.
• shape_predictor_68_face_landmarks.dat: Pre-trained dlib landmark model.
• requirements.txt: Python package dependencies.

---

📄 License

This project is licensed under the MIT License.