COVID Sentine - AI COVID19 Solution Compliance
This is an open source and cross platform project which serves as solution for COVID 19 pandemic.
Repository Video ▶️Covid-Sentine
AI Based Mask Detection System
It is used to detect whether people are wearing the masks or not.
This snippet is based on a trained neural network (using transfer learning) on custom data set. This is able to detect people with and without face mask using object detection technique
Motivation
Prevention is better than cure
✔WHO advises to wear mask and prevent touching Nose, Face and Eyes.
✔There is no vaccine till now, and wearing masks is one of the only way to get rid of this virus.
✔Prevention is the only way to control spread of Novel Corona Virus and will lead of flattening of curve.
Unique Selling Point (Mask Detection System)
✔Green: mask Red: no_mask : with their detection probabilities on top of each bounding box.
✔This system is highly scalable and can detect multiple streams in optimal time.
✔This has great use in surveillance and security purpose to find out who all are following the rules.
How to install
It is a software that can take feed from existing CCTV/IP cameras and show the results.
VideoLink as a sample output for my mask detection system
Use Cases
✔ Can be used at hospitals
✔ Can be used at Railway Stations
✔ Can be used at Crowded Areas
✔ Can be used at Market Places and Malls
AI Based Social Distancing Monitoring System
It is used to detect whether people are following the social distancing or not
Motivation
✔ WHO advises to maintain social distancing.
✔ Minimum permissible distance is 1 metre (3 feet).
✔ Prevention is the only way to control spread of Novel Corona Virus and will lead of flattening of curve.
Unique Selling Point (Social Distancing System)
✔ This is our AI Product to check if commuters in the monitored area are following Social Distancing or not.
✔ Red: Alert, Green: Distancing maintained
✔ Contrary to others, this system is highly scalable and can detect multiple streams in optimal time.
✔ The USP is that distance between two people is independent of their heights as well as it can detect upto 1000 people in a single frame.
✔ This system is based on Deep Learning and not on Conventional Computer Vision
✔ Can be deployed on edge and cloud both.
VideoLink as a sample output for social distancing system
Use Cases
✔ Can be used at hospitals
✔ Can be used at Railway Stations
✔ Can be used at Crowded Areas
✔ Can be used at Market Places and Malls
AI Based Redness of Eye Detection
This snippet of software will aim at detecting the people who are having conjunctivitis in their eyes
Motivation
✔ Conjunctivitis(Red Eye) is considered as the symptom of Covid19 as per the guidelines of World Health Organization.
.
✔ Our software detects red eyes and raises alert for probability of person being infected.
How to install
It is a software that can take feed from existing cameras and show the results.
AI Based Fever Detection
This snippet of software will aim at detecting the people who have high body temperatures.
Motivation
✔ We have considered the Human Face as Region of Interest to calculate the thermal temperature
✔ Our system will couple AI Based ROI Detection + Temperature Calculation in ROI
✔ Things to look for: If a person carries coffee or any hot food item, it won’t be considered for temperature calculation.
✔ Our software detects red eyes and raises alert for probability of person being infected.
Project Demo
🚀 Installation
It is a software that can take feed from existing CCTV/IP cameras and show the results.
This system can be installed and is tested on Jetson Nano. The cost is Jetson nano is 99$.
We combined all the above mentioned solutions together with a Graphical User Interface.
To set up on your local: 1. Clone the repo
$ git clone https://github.com//Sonali210/Covid-Sentine.git
- Change your directory to the cloned repo and create a Python virtual environment named 'test'
$ mkvirtualenv test
- Now, run the following command in your Terminal/Command Prompt to install the libraries required
$ pip3 install -r requirements.txt
Working
Open terminal. Go into the cloned project directory folder and type the following command:
$ python covid.py
Owner
Made by Gaurav Singh , Sarthak Srivastava and Sonali
License
Apache License 2.0
COVID Sentine
Demo link: https://youtu.be/YqOERi1Jjk8
Added face mask detection to check if the person is wearing a mask or not.
Project created by Sonali