Criminal Activity Detection System

Team Members

• Mayur Dehade
• Anjali Barde
• Dipak Bhide
• Milind Ghegadmal

Project Overview

The Criminal Activity Detection System is our final year engineering project aimed at enhancing security by detecting criminal activities and weapons in real-time. The system uses deep learning algorithms to analyze live video feeds or uploaded videos, identifying suspicious activities or weapons and notifying the appropriate authorities.

Project Features

• Real-Time Activity Detection: Uses Convolutional Neural Networks (CNN) to detect criminal activities as they happen.
• Weapon Detection: YOLO v7 algorithm identifies weapons in real-time, providing critical information to prevent potential threats.
• User Authentication: A secure login system that stores user credentials in an SQLite3 database.
• Live Camera Feed: Allows users to monitor live video feeds for real-time detection.
• Video Upload: Users can upload pre-recorded videos for analysis.
• Alert System: Sends text messages to designated individuals when suspicious activities or weapons are detected.

Project Modules

1. User Authentication Module:

   • Manages user registration and login.
   • Stores user credentials securely in the SQLite3 database.

2. Live Camera Module:

   • Activates the live camera feed for real-time activity and weapon detection.
   • Displays results of detection on the user interface.

3. Video Upload Module:

   • Allows users to upload pre-recorded videos.
   • Processes the video to detect any criminal activities or weapons.

4. Detection and Alert Module:

   • Runs the trained CNN and YOLO v7 models to detect criminal activities and weapons.
   • Sends a text message alert to the respective person upon detection.

5. Notification Module:

   • Integrates with a messaging service like Twilio to send real-time alerts.
   • Provides detailed information about the detected activity or weapon in the alert message.

How It Works

1. User Login: The user logs into the system using their credentials.
2. Choose an Option:
   • Start Live Camera: Activates the camera and starts real-time detection of activities and weapons.
   • Upload Video: Allows the user to upload an existing video for analysis.
3. Detection Process:
   • The CNN model detects criminal activities.
   • YOLO v7 detects the presence of weapons.
4. Notification: Upon detection, the system sends a text message to the designated person, providing details about the detected activity or weapon.

Project Specifications

• Programming Language: Python
• Framework: Django
• Database: SQLite3 (used for storing user login information)
• Deep Learning Algorithms:
  • CNN (Convolutional Neural Network): Detects criminal activities in real-time.
  • YOLO v7 (You Only Look Once, Version 7): Detects weapons in real-time.
• Training Data: The models are trained on thousands of images to achieve high accuracy.

Tech Stack

• Frontend: Django

• Backend: Python

• Database: SQLite3

• Deep Learning: TensorFlow, Keras (for CNN and YOLO v7)

• Communication: Twilio for sending text messages

  Installation and Setup

1. Clone the repository from GitHub:

   git clone https://github.com/mayurdehade/Criminal-Activity-Detection.git

2. Navigate to the project directory:

   cd Criminal-Activity-Detection

3. Install the required dependencies:

   pip install -r requirements.txt

4. Run the Django server:

   python manage.py runserver

5. Access the application in your web browser at http://localhost:8000.

Usage

• Log in to the system using your credentials.
• Choose to either start the live camera or upload a video for analysis.
• View real-time detection results or analysis of the uploaded video.
• Receive text alerts if any criminal activity or weapons are detected.