Theft Vehicle Detection & Tracking System

Features

• Vehicle Detection using YOLOv7
• Attribute Recognition: Class, color, brand, license plate
• Multi-Camera Tracking using turning patterns
• Google Maps API integration for route and address display
• MySQL backend for detection and metadata
• User-friendly Web Interface

Technologies Used

Area	Tech Stack
Frontend	HTML, CSS, JavaScript
Backend	Python (Flask)
Machine Learning	YOLOv7, OpenCV, Tesseract OCR
Database	MySQL
APIs	Google Maps Geocode & Route APIs

How It Works

1. User Input: Enter vehicle details or upload an image.
2. AI Detection: YOLOv7 detects vehicle class, color, brand, and license plate.
3. Tracking: Tracks turning patterns across adjacent cameras.
4. Location: Shows last seen address and route on Google Maps.
5. Output: Returns video snippet, CSV log, and live notifications.

Getting Started

Prerequisites

• Python 3.8+
• MySQL Server

Steps

git clone https://github.com/yourusername/theft-vehicle-detection.git
cd theft-vehicle-detection
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate
pip install -r requirements.txt

MySQL Setup

• Create a database named vehicle_tracking
• Import schema from /db/schema.sql

Run the App

python app.py

Project Structure

theft-vehicle-detection/
├── models/
├── static/
├── templates/
├── utils/
├── db/
├── app.py
├── config.py
├── requirements.txt
└── README.html

Sample Output

• Detection Accuracy: 85–90%
• Address: XYZ St, Bangalore, Karnataka
• Output: timestamp overlay, CSV log