🎯 Key Highlights

• 🧠 Advanced AI Models: YOLO11x, YOLOv11l-face, HRNetV2, MediaPipe
• ⚡ Full GPU Acceleration: CUDA optimization for real-time performance
• 📱 IoT Integration: ESP32 hardware alerts with Blynk app control
• 🎨 Professional Architecture: Modular, scalable, and maintainable codebase
• 🔊 Multi-Modal Alerts: Visual LED alerts, audio warnings, and mobile notifications

🚀 Features Overview

🏗️ System Architecture

┌─────────────────────────────────────────────────────────────────┐
│                        Driver Safety Guard 2.0                 │
├─────────────────────────────────────────────────────────────────┤
│  📹 Camera Input                                               │
│     └── Real-time video stream processing                      │
├─────────────────────────────────────────────────────────────────┤
│  🧠 AI Detection Engine                                        │
│     ├── 📱 Phone Detection (YOLO11x)                          │
│     ├── 👤 Face Detection (YOLOv11l-face)                     │
│     ├── 👁️ Drowsiness Detection (MediaPipe)                  │
│     └── 👀 Gaze Direction Tracking (MediaPipe)                │
├─────────────────────────────────────────────────────────────────┤
│  🎯 Tracking & Feature Extraction                             │
│     ├── 🔍 Multi-object Tracking                              │
│     └── 🧬 HRNetV2 Feature Extraction                         │
├─────────────────────────────────────────────────────────────────┤
│  📡 Communication Layer                                        │
│     ├── 🔊 Text-to-Speech Alerts                              │
│     ├── 📱 ESP32 HTTP Communication                           │
│     └── ☁️ Blynk IoT Platform                                 │
├─────────────────────────────────────────────────────────────────┤
│  🛠️ Hardware Integration                                       │
│     ├── 💡 RGB LED Alert System                               │
│     ├── 🔔 Audio Buzzer Alerts                                │
│     ├── 🚨 MQ2 Smoke Detection                                │
│     └── 🌡️ Temperature Monitoring                             │
└─────────────────────────────────────────────────────────────────┘

📋 Detailed Feature Breakdown

🧠 AI-Powered Detection Systems

📱 Phone Usage Detection

• Model: YOLO11x (State-of-the-art object detection)
• Capabilities: Real-time phone detection with high accuracy
• Performance: GPU-optimized for 30+ FPS processing
• Smart Logic: Confirms phone usage with person detection

👁️ Drowsiness & Eye State Monitoring

• Technology: MediaPipe Face Mesh + Custom algorithms
• Features:
  • Eye aspect ratio (EAR) calculation
  • Blink pattern analysis
  • Continuous drowsiness scoring
  • Customizable sensitivity thresholds

👀 Gaze Direction Tracking

• Method: Facial landmark analysis with 468 key points
• Tracking: Left/right gaze detection with duration monitoring
• Alerts: Triggered after 3+ seconds of sustained off-road gaze
• Accuracy: Sub-degree precision for gaze angle estimation

🚨 Smoke Detection

• Hardware: MQ2 gas sensor with ESP32 processing
• Features: Adjustable threshold via mobile app
• Integration: Real-time air quality monitoring
• Alerts: Immediate visual and audio notifications

🔧 Hardware Integration

🎛️ ESP32 IoT Controller

• Connectivity: WiFi-enabled with web interface
• API Endpoints: RESTful HTTP communication
• Real-time Control: Instant alert triggering and status updates
• Web Dashboard: Browser-based monitoring and configuration

💡 Smart LED Alert System

🔴 Red LED    → Distracted driving detected
🟡 Yellow LED → Drowsiness warning
🔵 Blue LED   → Phone usage alert
🟠 Orange LED → Smoking detected
🟢 Green LED  → System normal/idle

🔊 Multi-Modal Alert System

• Audio Alerts: Voice warnings with customizable messages
• Visual Alerts: Color-coded LED indicators
• Mobile Notifications: Push alerts via Blynk app
• Progressive Escalation: Increasing urgency based on behavior persistence

📦 Installation & Setup

🖥️ System Requirements

Minimum Requirements

• OS: Windows 10/11, Linux (Ubuntu 18.04+), macOS 10.15+
• CPU: Intel i5 / AMD Ryzen 5 (4+ cores)
• RAM: 8 GB
• Storage: 5 GB free space
• Camera: USB webcam or integrated camera
• Python: 3.8 or newer

Recommended Configuration

• CPU: Intel i7 / AMD Ryzen 7 (8+ cores)
• GPU: NVIDIA RTX 3060 or better (8+ GB VRAM)
• RAM: 16 GB
• Storage: 10 GB SSD space
• Camera: High-resolution USB camera (1080p+)
• CUDA: 11.8+ for GPU acceleration

🚀 Quick Start Installation

1. Clone the Repository

git clone https://github.com/menuka400/Driver-Safety-Guard-2.0.git
cd Driver-Safety-Guard-2.0

2. Create Virtual Environment

# Windows (PowerShell)
python -m venv venv
.\venv\Scripts\Activate.ps1

# Linux/macOS
python3 -m venv venv
source venv/bin/activate

3. Install Dependencies

# Install from setup.py (recommended)
pip install -e .

# Or install manually
pip install -r requirements.txt

4. Download AI Models

python utils/download_models.py

5. Configure the System

# Copy and edit configuration
cp config/default_config.json config/your_config.json
# Edit your_config.json with your preferences

6. Run the System

python main.py

🔧 ESP32 Hardware Setup

1. Hardware Requirements

• ESP32 Development Board
• MQ2 Gas Sensor Module
• DS18B20 Temperature Sensor
• RGB LEDs (5x for different alerts)
• Buzzer Module
• Resistors and connecting wires

2. Arduino IDE Setup

# Install required libraries:
# - WiFi
# - WebServer
# - BlynkSimpleEsp32
# - OneWire
# - DallasTemperature
# - ESPmDNS

3. Upload ESP32 Code

# Open arduino/main/main.ino in Arduino IDE
# Configure WiFi credentials and Blynk token
# Upload to ESP32 board

4. Network Configuration

# Find ESP32 IP address from serial monitor
# Update config/your_config.json with ESP32 IP
# Test connection: ping YOUR_ESP32_IP

⚙️ Configuration & Customization

📝 Configuration File Structure

{
  "detection": {
    "confidence_threshold": 0.7,
    "phone_detection_enabled": true,
    "face_detection_enabled": true,
    "drowsiness_detection_enabled": true,
    "gaze_tracking_enabled": true
  },
  "gpu": {
    "enabled": true,
    "device": "auto",
    "mixed_precision": true
  },
  "esp32": {
    "ip": "192.168.1.100",
    "port": 80,
    "enabled": true
  },
  "alerts": {
    "tts_enabled": true,
    "voice_rate": 150,
    "volume": 0.8,
    "gaze_threshold_seconds": 3.0
  }
}

🎛️ Customizable Parameters

🎮 Usage Guide

🖥️ Running the Main System

Basic Usage

# Start with default configuration
python main.py

# Use custom configuration
python main.py --config config/your_config.json

# Enable debug mode
python main.py --debug

# Disable GPU acceleration
python main.py --no-gpu

Command Line Options

python main.py [OPTIONS]

Options:
  --config PATH     Configuration file path
  --debug          Enable debug logging
  --no-gpu         Disable GPU acceleration
  --camera-id INT  Camera device ID (default: 0)
  --esp32-ip IP    ESP32 device IP address
  --help           Show help message

📱 Blynk Mobile App Setup

1. Install Blynk App

• Download from App Store or Google Play
• Create account and new project
• Use template ID: TMPL657qAruym

2. Widget Configuration

• V0: System ON button
• V1: System OFF button
• V2: Status display
• V3: Smoke level gauge
• V13: Smoke threshold slider
• V14: Temperature display

3. Get Authorization Token

# Copy auth token from Blynk app
# Update arduino/main/main.ino with your token
# Re-upload ESP32 firmware

🔧 ESP32 Web Interface

Access the ESP32 web dashboard at: `http://YOUR_E