Project Vision

Introduction

Surveillance means safety. Despite so much surveillance, the crime rates are increasing. The problem lies in the fact that inspecting several hundreds and thousands of videos is very laborous and time-intensive task. Only after a crime is committed, these footages are inspected to find the crime scenes evidence, which proves to be highly inefficient considering the massive amount of footages. This manual task provides evidence in court but is rarely used to prevent crime or react to it in real-time.

🔭   About this Project

In this project, we aim to develop a real-time crime detection technology that can be integrated with any security system, to ensure public safety through visual crowd surveillance.
We have devised a complete software solution for safety and surveillance where we convert these CCTVs from an evidence collection to a crime prevention and detection tool to ensure safety and security.
This real-time crime detection technology is integrated with security systems and Desktop Application to get push-in notifications in case any suspicious activity is discovered. The system will detect and simultaneously alert any violent activity captured by the camera.
The technologies encompasses domains of Computer Vision and Deep Learning models like CNN, R-CNN, LSTM. The dataset consists of 500+ videos scraped off internet and categorized into violent and non-violent activities.

🛠  Tech Stack

       
       

💼   Testing

Our source code can be tested and contributed to using the following steps:

1. Website

Execute the following sequence of code to start contributing and testing:

• Open the directory labelled Website using the command

  cd Website

• Install all the dependencies using

  pip3 install -r requirements.txt

• Run the migrations using the following commands

  python manage.py makemigrations

  python manage.py migrate

• Run the Django server by

  python manage.py runserver

Note : In case you encounter errors during migrations, make sure that you have access rights to db.sqlite3 file. You can use the following command to rectify permission denied error

chown *username* db.sqlite3

Run the migrations again to continue.

2. Desktop App

Being built on top of Node.js, it is fairly easy to set-up the environment. Follow the steps to safely configure the system

• From the parent folder directory, execute the following command

  npm install

• Once finished, the application can be run using

  npm start

• The desktop app has two modes of operation: Offline and Online. By default the app function in Online mode. To make it work in Offline mode, use the menu button provided on top of UI and select Offline mode.
  
  

✍️   Current Progress

We tried and tested various architectures for this project. A brief summary and stats for each is given below:

1. Customised CNN Model: The customised CNN model with optimised parameters performed well on the training and sufficiently good in the test dataset. The architecture is currently being used by the website and test.ipynb file.

   • Accuracy on training: 0.7990
   • Loss on training: 0.4292
   • Val_loss: 0.7605
   • Val_accuracy: 0.6680
   • Optimizer: Adam
   • Loss: BinaryCrossEntropy
   • Epochs: 50
   • steps_per_epoch: 50
     
     

   

   
   
   

2. VGG Net: VGGNet Architecture displayed an accuracy of 60% on training and 55% on testing dataset.

   • Testing Accuracy: 0.5513
     
     

   

   
   
   

3. AlexNet: AlexNet showed accuracy of 57% on training and a similar accuracy on the testing dataset.

   • Testing Accuracy: 0.5729
     
     

   

   
   
   

4. Inception+CustomisedCNN: Using transfer learning of Inception Architecture and passing it to CustomisedCNN trained the model with satisfactory results.

   • Training_accuracy: 89%
   • Validation_accuracy: 76%
   • Epochs: 30
   • Steps per Epoch: 100
   • Optimizer: RMSprop with LR 0.0001
   • Loss: Binarycrossentropy
     
     

   

   
   

⚙️   Ultimate Objective

The model uses the technique of Multiple Object Detection with Localization tracking the movement of people and then categorizing it into violent or non-violent behavior. Our proposed solution will ensure Public Safety and security without any human toil with an instant alert to the concerned authorities. Through Crowd Monitoring and Behavioral Analysis our solution aims to evoke the sense of security in men and women. The constant rise of criminal activities, their unexpectedness, and scope of harm that can be inflicted can be exponentially reduced through our proposed system.

Link to current protoytpe : Click Here

Link to Desktop Apps

Platform	Link
Windows (.exe)	Link
Linux (.AppImage)	Link
Mac (.dmg)	Link

Custom Build

If you want to create the desktop app for a different platform or a different extension like .deb, .rpm, use the following command from parent folder

electron-forge make