Rapid Verification of Covid Personnel

Project Title

Rapid Verification of Covid Personnel

Project Area of Specialization Information & Communication TechnologyProject Summary

‘Rapid COVID-19 Vaccination of Vaccinated Personnel’, which aims to provide a secure method to check a person’s vaccination status. In past years, the COVID-19 Virus has taken center stage in discussions about daily survival around the world. Many lives have been lost and several people have suffered financially. As in 2021, pharmaceutical companies around the world have introduced the COVID-19 vaccine and every country’s ruling authority has taken steps to make the vaccination mandatory for the public. The method of checking in Pakistan has mainly been by eye, i.e. a person who has gotten vaccinated will show their vaccination certificate to the worker who is appointed to check for the certificate at the public place. The problem with this method is that the worker might not be aware of the legitimacy of the certificate, and neither does a system exist to check it in real-time Hence, a solution is proposed based on Facial Recognition by the help of Edge Computing.

The project will be using the facial recognition method to identify the person. During this procedure, a facial image will be captured of a person in front of a camera. Then it is transferred to the first edge computing device which contains a microcontroller where the facial recognition software reads the geometry of the person’s face. Key factors include the distance between the person’s eyes, the depth of your eye sockets, the distance from forehead to chin, the shape of your cheekbones, and the contour of the lips, ears, and chin. The aim is to identify the facial landmarks that are key to distinguishing your face, as the image is processed and a unique code is generated from the picture. In this way, the first objective will be completed in this phase.

Moreover, edge computing technology is used which ensures each data processing component contains its device which maximizes the processing speed so the delay can should be minimized and working of the project becomes rapid. To establish the technique of edge computing we have placed a device at entry-level with components such as a camera, display and scanner machine. The second device is placed near the servers where the information of members is kept in the database.

Project Objectives

  Project Objectives:

• To generate a real-time Image of the person for facial recognition.
• To interface the camera and/or card reader with the edge computing module.
• To interface edge computing module with the database.
• To achieve real-time and rapid results by using Edge Computing at the server end.
• To display the achieved results on the LCD.

Project Implementation Method

 Implementation Details:

As explained earlier we have divided our system in two parts uptill now for implementation, First one includes the Facial Recognition algorithm and second one includes the Back haul link based on edge computing.

Facial Recognition

The technique used in the facial recognition is the well known Viola Jones Object Detection Framework. The details of the features of this framework are highlighted below:

• Haar Feature Selection
• Adaboost Training
• Cascading Classifiers

A Haar-like feature considers adjacent rectangular regions at a specific location in a detection window, sums up the pixel intensities in each region and calculates the difference between these sums. This difference is then used to categorize subsections of an image.

Adaboost Training, or Adaptive Boost Training is a machine learning algorithm which is based on the combined functionality of many weak learners in an algorithm, and the outputs of these weak learners are combined to produce a strong final output. This algorithm constructs a “strong” classifier as a linear combination of weighted simple “weak” classifiers. The reason its used in this Viola Jones detector is because it greatly reduces the time delay required to train the classification model.

Feature Detection and Extraction

For the implementation, MATLAB’s webcam tool is used to detect realtime images of faces and perform facial recognition on them. A database of 120 images is created using the web camera, which is further sorted into 4 folders, each folder corresponding to one person and containing 30 images each. The CascadeObjectDetector is the MATLAB function which is based on the Viola Jones Object Detector, and is used to detect the faces through the web camera.

Image Classification

Next, this dataset is loaded into a file containing the code for the learning algorithm, which consists of the Alexnet tool. The tool is trained to recognize the images of 4 persons and achieves an accuracy of 100% after 50 iterations. This creates the classification model used to identify real-time images.

Real Time Facial Recognition

Finally, after the classification model has been created, the web camera is used again to identify real time images of the person and the name of the person identified is presented in the output window.

 For the implementation of Back Haul link:

Benefits of the Project

The project will help the community with permission for access to a specific organization. As most organizations require a vaccination certificate for entry from gates. Our project will be made entry easy with the help of contactless entry. Benefits of this project include:

• Contactless Entry to organization
• Rapid Verification
• Efficient edge devices save power

The project will help multiple organizations such as schools, offices, shopping malls, etc.

Technical Details of Final Deliverable

We have developed this project on hardware using edge devices, cameras, card readers, servo motors, etc. Most of the equipment details are listed below:

EDGE DEVICES

 We have used mini CPUs as our edge devices to handle processing power at both edges and generate the results. For mini CPU we have decided to use AMD RYZEN 7 based processor for fast processing along with a RADEON graphic card to handle image conversions at a high processing level.

 

WEB CAMERA

A Web Camera is a compact digital camera device to broadcasts real-time images or videos. It’s not very different from a digital camera; it also converts the real picture into a digital picture through image sensors and its circuitry. Unlike a digital camera, a web camera does not have any built-in memory space because it immediately transfers images to the computer. Thus, a webcam has a USB port attached to its back. Face recognition will be possible with the webcam.

CARD READER

A card reader, also known as a magnetic stripe reader, reads the data stored on the magnetic stripe of your card, which includes personal information like your name, address, and ID card number. A card reader is a device that reads and decodes the code. In our project, a card reader will read the information on the ID card and determine whether or not the individual who owns the ID card has been vaccinated.

LCD

We have used a 20 inch LCD to interact with users so they can place their face at better angle for picture and to show them output results on the screen for their vaccinated status.

SERVO MOTOR

 We have used a servo motor at the gate entry level which will control the barrier and let the people block or allow them to enter the premises of the organizations.

Final Deliverable of the Project HW/SW integrated systemCore Industry ITOther Industries Education , Health , Security Core Technology Cloud InfrastructureOther Technologies Artificial Intelligence(AI), Internet of Things (IoT)Sustainable Development Goals Sustainable Cities and CommunitiesRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	80000
AMD Ryzen mini CPU	Equipment	2	26000	52000
Camera	Equipment	1	10500	10500
LCD	Equipment	1	7500	7500
Barcode Reader	Miscellaneous	1	7000	7000
Servo Motor	Miscellaneous	1	3000	3000