Skip to content

Latest commit

 

History

History
62 lines (42 loc) · 1.83 KB

README.md

File metadata and controls

62 lines (42 loc) · 1.83 KB

AutoGuard 🛡️🚗

AutoGuard is an Android application designed to detect distractions and drowsiness while driving, promoting safer driving habits.


🛠️ Technologies Used

  • MediaPipe: Real-time facemesh implementation for facial detection.

  • MPAndroidChart: Charting library for visualizing distraction statistics.

  • Java: Native Android development.

    Tecnologies


🚀 Key Features

  • Facial Distraction Detection: Analyzes the driver’s attention in real time.
  • Smart Notifications: Alerts based on the severity of distractions.
  • Detailed Graphs: Visualizes statistics by hours and days.
  • User-Friendly Interface: Intuitive design with accessible buttons.

📸 Screenshots

Real-Time Monitoring

Monitoring 2

Detailed Statistics

Statistics 1 Statistics 2

Project presentation

Presentation 1 Presentation 2


📈 How It Works

  1. Real-Time Facemesh: Uses MediaPipe's algorithm to map key facial features.
  2. Distraction Detection: Calculates parameters such as eye opening and movements to evaluate attention levels.
  3. Data Visualization: Collected statistics are displayed through detailed graphs.

🏆 Final Grade

This project was graded 9.3/10, showcasing its effectiveness and innovation.