Software Engineering Group Project: Group 14 - DETECTING HEARTBEAT VIA COLOUR AND/OR MOTION MAGNIFICATION

Team Members: Imogen Green, Utkarsh Gupta, Matthew Grouse, Dishant Tekwani, Max Cunningham, Victoria Fabijaniak, Niall Connolly . Demonstrator: Matt Murtagh.
Client: Serkan Ozkul and Andrew Quirk - Microsoft .

Build on localhost

1. Clone the repo:

git clone https://github.com/Utkarsh803/Heartbeat_Detection

2. Install dependencies

npm install
pip install -r requirements.txt
cd client && npm install

May need manual installing:
cd client
npm i devextreme-react
npm i react-circular-input
npm i --save react-signals-plot

Authentication and database required to be set up, but are not required for basic functionality
For Microsoft Authentication Library MSAL, add information in Heartbeat_Detection/client/src/authConfig.js
See here: https://docs.microsoft.com/en-us/azure/active-directory/develop/tutorial-v2-react

For Database, add information in Heartbeat_Detection/database.py
https://docs.microsoft.com/en-us/azure/azure-sql/virtual-machines/windows/sql-vm-create-portal-quickstart

3. Start the server

cd Heartbeat_Detection
python endpointbase.py (without the database)
python endPoint.py (with the database)

cd client
npm start

App Functionality

Description

This application aims to detect heartbeat via Frontal Face Video and provides accurate results with error margin of <= 4 BPM* (tested on 8 subjects) in suitable environment and light.

Video Demonstration:

https://youtu.be/Tzwx9WraMEw

Software Engineering Project - Group 14

• Imogen Green ([email protected] - 17326096)
• Utkarsh Gupta ([email protected] - 19312536)
• Matthew Grouse ([email protected] - 19335171)
• Max Cunningham([email protected] - 20333171)
• Niall Connolly([email protected] - 20332921)
• Victoria Fabijaniak([email protected] -20332659)
• Dishant Tekwani([email protected] -20309212)

References

• Real Time Heart Rate Monitoring From Facial RGB Color Video Using Webcam by H. Rahman, M.U. Ahmed, S. Begum, P. Funk
• Non-contact, automated cardiac pulse measurements using video imaging and blind source separation by Ming-Zher Poh, Daniel J. McDuff, and Rosalind W. Picard

*readings taken after input signal stabilizes