Face Mask and Social Distance Detector in covid-19

Project Title

Project Area of Specialization

Computer Science

Project Summary

During this year we are encountering that COVID-19 is a very spreading infection, in practically entire world there are many patients. Whole of the world is creating rules and taking safety measures to be safe from this disease. In our country for safety measures our government of Pakistan has applied a standard of wearing mask and keeping up a separation between one another at public spots which is said as social distancing. As we have seen police, other government departments and a portion of the private department are working day and night to ensure that individuals adhere to the standard of wearing masks and keep up social distancing. This venture is programmed to help those individuals by essentially utilizing it on observation cameras or CCTV in structures, workplaces, emergency clinic and others. Utilizing this will diminish the human work and it will be simpler to identify those individuals who are not wearing masks and keeping up social distancing in public and work zones. Using this sort of system will help both the people who are working day and night to make people follow the SOP’s of social distancing and wearing a mask. It uses the real time image through camera and identifies if the person is wearing mask and if there are two or more persons there is a safety distance or not, we can attach it to the surveillance cameras, public cameras, CCTV of a building, office, shopping malls and other places which holds daily life activities.

Project Objectives

This project will allow the users to identify a person without mask and social distance between two persons through camera.

. Its objectives are:

Mask detection make it possible to identify the person who is not wearing a mask.
Detecting the distance between two persons to identify if they are in a safe distance or not.

Base objective is to decrease the spread of covid-19 infection that is spreading worldwide in an enormous speed
To decrease the human labor and work through the camera systems like CCTV and surveillance cameras in any area.
Tough job of humans checking and identifying those without masks in public areas and work zones, they also get infected because of the physical and less distance when catching any person who is infected and not wearing mask.

Project Implementation Method

The two main methodologies which we further adjusted for our framework working are model based face recognition and model biased person body recognition. By all means we wanted our camera to detect the mask on a person and calculate the distance between two persons to identify if they are in safe distance or not. So first of all, we will make train a model through different images of people wearing masks to make sure it has a good accuracy percentage we will try all the images of male, female and kids. And of all ages and all genders that will emphasize the accuracy rate of the detector when detecting mask on a person face. As we need to detect the social distance through real time image at the same time we will use a person detecting models there are several available but we will further edit it and then program it with a distance formula through arrays to calculate the distance and check if they are in safe distance or not.

Training a model for mask detection through images of all genders, of all ages and of different appearances.
Using a person detector model and further editing it for our own frame work.
Applying distance formula on persons distance detector.
Euclidian distance Formula.
Convolutional neural network used for training the model.

Benefits of the Project

The mask detection and social distancing is project is basically to develop for maintain the COVID 19 SOP's. We are a third world country and, in our country, we are facing lots of death in daily routine and we cannot afford this situation. So we have to follow the SOP and maintain our criteria to deal with this situation. So that is why we are working on this project to handle this situation for saving the country losses. And we have a lot of more issues like our markets and all the business are in the loss so we have to face all these issues so that’s why we decide to make this type of project that deals with this situation

Technical Details of Final Deliverable

This program is actually a face mask and social distancing detector it will identify if the person is wearing a face mask or not and if there is a safe distance between them or not. It is a desktop application that will run and work through camera and it will recognize mask and social distance it basically requires a computer and a webcam to run through but to run it and execute it properly there libraries to add in python code and there are models to do so the frame work works properly. By identifying a person’s face first then a mask on it then the social distance between two persons on real time image or video it will show a real time running video on client side in which it will mention if the person is wearing mask or not. If there are more than two, they are in a safe distance or not. Data sets are required to train the model they should be image data set of people wearing a mask which we have already made. After this a proper code to identify the real time processing image as per required needs.

Final Deliverable of the Project

Software System

Core Industry

Other Industries

Core Technology

Artificial Intelligence(AI)

Other Technologies

Others

Sustainable Development Goals

Good Health and Well-Being for People

Required Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
Report	Miscellaneous	12	500	6000
HDMI Wire	Equipment	2	2000	4000
Hard Disk	Equipment	3	8000	24000
RAM	Equipment	1	8000	8000
Mouse	Equipment	1	2000	2000
GPU - NIVIDIA Toolkit	Equipment	2	16000	32000
			Total in (Rs)	76000