Team Tech Vanguard

Pawan kushwah

Rakshita K Biradar

Neeraj parmar

Ravi Rajpoot

Nirmal Mewada

Women Safety SOS https://github.com/RAVI-RAJPUT-UMATH/Women_Safety : A quick emergency response system for women's safety.

Signal Automation https://github.com/pawankushwahh/Signal_Automation : An AI-driven system for optimizing traffic signals based on real-time traffic density.

Traffic Rule Violation Detection and Notification System** (This repository): Automated detection of traffic rule violations using computer vision.

Video Demonstration https://drive.google.com/file/d/1GaEdtEzO_qE81oPrjO9-C9Vlkk_zl1CV/view?usp=sharing

Traffic Rule Violation Detection and Notification System

An automated system that detects traffic rule violations using computer vision, processes the violations, and sends notifications to both violators and traffic authorities.

Features

• Real-time Violation Detection: Uses YOLOv8 for detecting vehicles and identifying violations
• Multiple Violation Types: Detects various violations including:
  • No Helmet
  • Wrong Side Driving
  • Signal Jump
• Automated Notifications: Sends instant email notifications to:
  • Vehicle owners
  • Traffic authorities
• Violation Logging: Maintains detailed logs of all violations
• Fine Generation: Automatically calculates fines based on violation type

How It Works (the flow)

The project runs in two simple steps. They are connected through one CSV file:

 video.mp4
    │
    ▼
[ STEP 1: violation_detector.py ]
    - YOLO finds vehicles, EasyOCR reads number plates
    - decides if a rule was broken
    - writes each violation as a row in  data/violators_data.csv
    │
    ▼
 data/violators_data.csv   ← the link between the two steps
    │
    ▼
[ STEP 2: notification_system.py ]
    - reads each violation row
    - looks up the owner (vehicle_database.csv) and police zone (location_authorities.csv)
    - emails both, and saves a copy in notification_logs/

You can run both steps at once with main.py.

Prerequisites

1. Python 3.8 or higher
2. A Gmail account (or any SMTP provider) for sending emails
   • You can also run in simulation mode with no email account at all.

Installation

1. Clone the repository:

   git clone https://github.com/pawankushwahh/traffic-violation-detection.git
   cd traffic-violation-detection

2. Install required packages:

   pip install -r requirements.txt

3. Create your .env file (copy from .env.example) and fill in your values:

   SMTP_HOST=smtp.gmail.com
   SMTP_PORT=587
   SMTP_USER=your.email@gmail.com
   SMTP_PASSWORD=your_gmail_app_password
   FROM_EMAIL=your.email@gmail.com
   EMAIL_SIMULATION=true
   

   Gmail setup: Enable 2-Step Verification, then create an App Password and use that as SMTP_PASSWORD. Keep EMAIL_SIMULATION=true while testing so no real emails are sent.

Project Structure

Traffic-management-and-women-safety/
├── config.py                     # ALL settings live here (read this first!)
├── main.py                       # run the whole pipeline (step 1 + step 2)
├── violation_detector.py         # STEP 1: detect violations in a video
├── notification_system.py        # STEP 2: email owners and authorities
├── email_service.py              # helper that actually sends the emails
├── data/
│   ├── vehicle_database.csv      # owner info (name, phone, address, email)
│   ├── violators_data.csv        # detected violations (the link between steps)
│   └── location_authorities.csv  # police zone contact details
├── violations/                   # snapshot images of caught vehicles
├── notification_logs/            # a text copy of every email
├── .env.example                  # template for your .env file
└── requirements.txt              # project dependencies

Usage

Easiest: run everything at once

python main.py --video video.mp4

If you don't want a video window to open (e.g. on a server):

python main.py --video video.mp4 --no-display

Or run the two steps separately

1. Detect violations (writes rows into data/violators_data.csv):

   python violation_detector.py --video video.mp4

2. Send notifications for everything recorded so far:

   python notification_system.py

Simulation vs. real emails

• By default (EMAIL_SIMULATION=true), emails are only printed and saved to notification_logs/ — nothing is actually sent. Great for testing.
• To send for real, set EMAIL_SIMULATION=false in .env, or use the --send flag:

  python main.py --notify-only --send
  python notification_system.py --send

Data File Formats

1. data/vehicle_database.csv:

   Vehicle_Number,Owner_Name,Phone_Number,Address,Owner_Email
   UP32UV1111,Aditya Mishra,+91XXXXXXXXXX,"45 Indira Nagar, Lucknow",aditya.mishra@example.com
   

   (Addresses that contain a comma must be wrapped in "double quotes".)

2. data/violators_data.csv:

   Vehicle_Number,Violation_Type,Location,Violation_Time,Fine_Amount
   UP32UV1111,No Helmet,Hazratganj,2025-02-15 08:20:00,500
   

Notes / Known Limitations (for students)

• Helmet detection is a placeholder. The default yolov8n model was not trained on helmets, so has_helmet() in violation_detector.py currently always returns True. To make "No Helmet" work, plug in a helmet-trained YOLO model there.
• Detected plates must exist in vehicle_database.csv for an email to be sent. If a real-world plate is read that isn't in the database, the notification step logs a warning and skips it (this is expected).
• The sample rows already in data/violators_data.csv let you test the email step on their own, without running detection.

Notification Format

1. Violator Notification:

   Traffic Violation Notice
   Dear [Owner Name],
   Your vehicle ([Vehicle Number]) was detected violating traffic rules:
   Violation: [Violation Type]
   Location: [Location]
   Time: [Timestamp]
   Fine Amount: Rs. [Amount]
   

2. Authority Notification:

   New Traffic Violation Detected
   Vehicle Number: [Number]
   Violation Type: [Type]
   Location: [Location]
   Time: [Timestamp]
   Vehicle Owner: [Name]
   Owner Contact: [Phone]
   Fine Amount: Rs. [Amount]
   

Troubleshooting

1. No Notifications Being Sent:

   • Check if SMTP credentials are correctly set in .env
   • For Gmail, use an App Password (not your regular password)
   • Verify vehicle exists in database with a valid Owner_Email
   • Check network connectivity

2. Vehicle Not Found:

   • Ensure vehicle information is present in vehicle_database.csv
   • Check if vehicle number format matches database format

Contributing

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contact

mail - Kushwahpawan2005@gmail.com Project Link: https://github.com/pawankushwahh/traffic-violation-detection