Crime Detection System using Face Recognition

The Face Recognition Web Application for Crime Prevention is an application that utilizes machine learning technology to recognize faces and prevent crime.

Consists of 2 main components:

01) Machine Learning Model: is trained using labeled images and is capable of recognizing faces with 98%+ accuracy. Once the model is trained, it is exported as a JSON file and linked with the web application.
02) Web Application: is designed to work with the deployed machine learning model and can detect faces with 98%+ accuracy within 4 seconds. The application uses a camera to capture images of people in a given area, which are then processed by the machine learning model to recognize the faces.

Publication

The research was conducted from May 2022 - July 2022 and was associated with the Indian Journal of Applied Research (IJAR).
The research findings have been accepted for publication in the 1st September 2022 issue of the IJAR.
Link

This project represents a significant contribution to the field of machine learning and facial recognition technology, with potential applications in law enforcement and public safety. By accurately identifying suspects and tracking their movements, this system can help prevent crimes and improve public safety.

Architecture

System consists of following components

Machine Learning Model consists of 5 components:

Image Capture
Image Labelling
Model Training
Model Evaluation
Model Export

Web Application consists of 4 components:

User Interface
Server
Email Notification System
Database