Automated-Traffic-Accident-Detection

🚧 Automated Traffic Accident Detection and Alert System

This project is an AI-powered system for detecting road accidents in real-time and classifying their severity using the YOLOv8 deep learning model. It also integrates risk scoring, insurance policy recommendations, and alert mechanisms (email/SMS), all within a secure web application built using Flask and MongoDB.

📌 Features

✅ Accident detection using YOLOv8 object detection
🔥 Real-time webcam/image/video feed processing
📊 Severity classification: Minor, Moderate, Severe
📈 Risk scoring and dynamic policy suggestion
📤 Email/SMS notifications for severe cases
🧑‍💻 Admin dashboard to manage users and policies
🔒 User login/signup with secure session handling

🧠 Technologies Used

Technology	Purpose
YOLOv8	Real-time object detection
Flask	Backend framework
HTML/CSS	Frontend interface
Bootstrap	Responsive web design
MongoDB	NoSQL database for user/detection data
OpenCV	Image and video processing
SMTP/Twilio	Sending email/SMS alerts

🚀 Installation

Clone the repository
```bash git clone https://github.com/your-username/accident-detection-system.git cd accident-detection-system

2.Create a virtual environment (recommended) python -m venv venv source venv/bin/activate # On Windows: venv\Scripts\activate

3.Install dependencies pip install -r requirements.txt

4.Configure MongoDB & Environment Variables

Ensure MongoDB is running locally or use MongoDB Atlas.

Set up .env for sensitive data (API keys, DB URI, email credentials).

5.Train or Load YOLOv8 Model

Download or train a YOLOv8 model on your custom dataset.

Place best.pt in the model/ directory.

▶️ Running the Application python app.py Then open your browser and go to: http://127.0.0.1:5000/

📁 Project Structure accident-detection/ │ ├── static/ │ └── uploads/, css/ ├── templates/ │ └── *.html ├── model/ │ └── best.pt ├── app.py ├── Main.py ├── requirements.txt └── README.md

🧪 Sample Use-Cases 🚨 Detect accidents from uploaded images or live webcam

📤 Alert admin and users during severe accidents

🧾 View detection history and personal risk level

📉 Admin assigns insurance policies based on risk