3 IN ONE: SMART COVID-19 BASIC SOP CHECKING DEVICE

Project Title

3 IN ONE: SMART COVID-19 BASIC SOP CHECKING DEVICE

Project Area of Specialization Artificial IntelligenceProject Summary

Coronavirus has been spread all over the world and it’s a deadly virus. It has taken many lives throughout this year and its vaccine is under clinical trial. The only solution to prevent this virus is to take SOPs that are guided by the doctors and World Health Organization (WHO).

 For this reason, we are designing a device that will examine the basic SOPs of Covid-19 that include:

• Temperature Checking
• Mask Detection,
• Social Distancing 
• Automatic sanitization

We will use the MLX90614 IR Temperature sensor that could detect the temperature of a specific person without getting in contact with them. For mask detection, Convolution Neural Network (CNN) process is used for the structured deep learning process that plays an innovative push for a variety of applications focusing on computer vision and image-based applications. For sanitization, a simple transistor or MOSFET with an IR proximity sensor is used. The IR sensor will detect the presence of the hand.

References:

[1] https://circuitdigest.com/microcontroller-projects/ir-thermometer-using-arduino-and-ir-temperature-sensor   (accessed at 9 December 2020. 7.39 pm)

[2] Sammy V. Militante, Nanette V. Dionisio. “Real-Time Facemask Recognition with Alarm System using Deep Learning”, 2020 11th IEEE Control and System Graduate Research Colloquium (ICSGRC 2020), 8 August 2020, Shah Alam, Malaysia. 

[3]https://www.hackster.io/taifur/automatic-hand-sanitizer-dispenser-no-arduino-ce852c  (accessed at 10 December 2020. 1.40 pm)

[4] Aryan Mehta, Mohan Kshirsagar, Reetu Jain, Shekhar Jain. “Autonomous UV Sanitization Robot with Social Distancing, Body Temperature and Mask Detection Using Automatic Path Planning and Multi-Terrain Capabilities”, International Journal of Scientific Research & Engineering Trends Volume 6, Issue 5, Sept-Oct-2020, ISSN (Online): 2395-566X

[5] Shams, Shahbaz A., Abid Haleem, and Mohd Javaid. "Analyzing COVID-19 pandemic for unequal distribution of tests, identified cases, deaths, and fatality rates in the top 18 countries." Diabetes & Metabolic Syndrome: Clinical Research & Reviews 14.5 (2020): 953-961.

[6]https://covid19.who.int/?gclid=Cj0KCQiA2uHBRCCARIsAEeef3nunqF1xg5C5O1I8eBzk_Vc4Eq-WIrSYqsnimTrsNyXvq8hgrpVL6saAudFEALw_wcB (accessed at 15 December 5.56 pm)

Project Objectives

The main objective of this project is to carefully check the SOPs so that this virus could not spread further. For example, sometimes due to rush in offices, malls, restaurants the guards may skip some people and they would enter without examine themselves and there is a risk of spreading the virus. It is prompted by the World Health Organization (WHO) to keep up with the guidelines. This device will help in examining everyone where ever it is placed. The guidelines include normal fever, wearing masks, social distancing, and sanitizing hands. In sanitization the primary issue is the manner in which it is done, that is by actual touch. Contacting hand sanitizers with infected hands can spread the infection to the following individual. Wearing a mask is another important SOP, people usually forget to wear a mask or they don’t want to wear a mask despite knowing that it is for their own safety, Normal temperature is one of the signs that a person is healthy and does not have corona as its symptoms include high fever also. It will automatically detect the temperature and if it’s above 37?C or 98.6?F then the person has a fever. Maintaining social distance will refrain people from coming in close contact and the germs cannot be transmitted from one another.

Project Implementation Method

Project implementation comprises of the following steps:

• Temperature checking
• Mask detection
• Social distancing
• Sanitization 

• Temperature Checking:

For temperature checking, we will use the MLX90614 IR Temperature sensor that could detect the temperature of a specific person without getting in contact with them. It will work as a thermometer in the device [1].

• Mask Detection: 

 For mask detection, Convolution Neural Network (CNN) process is used for the structured deep learning process that plays an innovative push for a variety of applications focusing on computer vision and image-based applications. CNN is widely used for;

• Face recognition.
• Object recognition.
• Classification of images.

             

  Fig 1. Real-time Facemask Recognition Framework [2].

• Capturing Images:

Fig.1 shows the step of CNN and the first step for real-time mask detection is to capture images of the people with the mask and without the mask from a camera that will be installed in the device.

 

• Data Set Collection:

2nd step is a data set that is created by labeling the pictures that we capture with the mask and without the mask.

• Image Processing:

Image processing is a 3rd step where the captured images are enhanced specifically for the image features during processing. This process divides the images into specific categories;

•  The one with the mask. 
• The one without a mask.

• Feature Extraction:

Feature extraction is the 4th step. In this process, it will detect the mask on the face that the person is wearing or not. It will extract other parts except where it will detect the presence of a mask. This process is done by convolutional and pooling layers.

• Classification:

The last step is to classify the images on deep learning model so that it should be trained to detect the images with mask and without the mask [2]. Fig 2. Shows the expected result of mask detection:

       

                                                             Fig 2. Mask Detection.

• Social Distancing:

In this process, camera video feeds from the Network Video Recorder (NVR) are transferred utilizing RTSP afterward these filters are changed over to grayscale to improve speed and exactness and are sent to the model for additional processing inside raspberry pi4. It also includes the SSD MultiBox Detector, a neural network architecture that has already been prepared on a large collection of images such as ImageNet and PascalVOC for high-quality image classification. To compute the distance between two people first the distance of the individual from the camera is determined utilizing the triangle similarity technique. Fig. 3 shows the concept of social distancing. Red labels show the people who are not following the social distancing whereas green labels show the people who are maintaining the social distancing.

                       

                                                Fig 3. Concept of Social Distancing                                      

• Sanitization:

For this purpose, we will use a simple transistor or MOSFET with an IR proximity sensor. The IR sensor will detect the presence of the hand. As it will detect hand it will release the sanitizer [3].

Benefits of the Project

Through this project, people can prevent themselves from spreading Coronavirus. When there will be a decrease in Coronavirus, the death ratio of the world will decrease and the economy of the world will again restore back to normal. This prime objective of the project is to focus on the Good Health and  Well-Being of humanity

Technical Details of Final Deliverable

The final product is a device embedded with an MLX90614 Temperature sensor for temperature checking with an LED display. Then 5MP raspberry pi camera module v1.3 will be used for image capturing such that the detection of the mask would be analyzed and then it is trained on OpenCV software. 5V sanitization pump is used with an IR sensor to detect the presence of a hand. The proposed device design is shown in Fig 4:

                                      

                                                  Fig 4. Proposed device for checking basic SOPs

Final Deliverable of the Project HW/SW integrated systemCore Industry HealthOther IndustriesCore Technology Internet of Things (IoT)Other Technologies Clean TechSustainable Development Goals Good Health and Well-Being for PeopleRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	74000
IR Sensor	Equipment	3	100	300
BD140 PNP BJT	Equipment	10	5	50
5V water pump	Equipment	3	200	600
Lithium cell	Equipment	10	170	1700
2x 18650 Battery cell holder	Equipment	6	70	420
MLX90614 Temperature sensor	Equipment	4	2300	9200
Arduino Nano	Equipment	4	380	1520
I2C OLED display	Equipment	3	450	1350
Piezo buzzer	Equipment	3	20	60
TP4056 battery charging module	Equipment	4	70	280
APDs-9930 RGB infrared gesture sensor	Equipment	3	380	1140
5MP raspberry pi camera module v1.3	Equipment	4	700	2800
Raspberry Pi 4 1GB RAM	Equipment	3	8500	25500
Raspberry Pi 4 Case with Fan Fitting Raspberry pi 4 Casing Cover	Equipment	3	360	1080
5 inch touch screen hdmi LCD	Equipment	4	4500	18000
Travelling for Consultation	Miscellaneous	3	1000	3000
Casing of the device	Miscellaneous	3	2000	6000
Thesis Printing	Miscellaneous	5	200	1000