Claims:The scope of the invention is defined by the following claims:

Claim:
1. A system/method for screening and tracking of Covid’19 infected persons face scans using deep learning algorithms, said system/method comprising the steps of:
a) The system starts to take inputs from dual mode cameras such as thermal scanning (1) as well as face scan images (5).
b) Threshold value of thermal (2) is given to the decision-making system (3), to check the positive (8) and negative labels (4).
c) Based on the face scans inputs, the face scan detection/tracking system (6) using deep learning algorithms (7), to check the decision of input images.
2. As mentioned in claim 1, the invented system starts with dual mode camera inputs like thermal and face scans.
3. According to claim 1, based on the thermal inputs, validate the thermal threshold values to check the decision system. If the system leads to positive then the patient has to take medical observations. If its negative values, then no need of medical observations.
4. According to claim 1, based on the face scan images input, it will send to face detection/tracking systems using deep learning algorithms with the respect to indicative features.
5. According to claim 1, the algorithm input is given to decision detection system, the results is depending on positive and negative values of decision-making system. , Description:SCREENING AND TRACKING OF COVID-19 INFECTED PERSONS USING FACE SCANS
Field of Invention
The present invention relates to screening and tracking of Covid-19 infected persons using face scans by deep learning algorithms.
The Objectives of this Invention
The major objective of this invention is to assist the authorities to automate the screening and tracking process not to replace the existing medical testing procedures.
Background of the Invention
In recent years, the Coronavirus (COVID-19) is the greatest challenge of mankind in the 21st century. The development of the disease, its transmission, and the increased mortality in a number of countries, make it imperative to develop treatment, but also to protect health care and society from the transmission of the disease. The COVID-19 pandemic has triggered an urgent need to contribute to the fight against an immense threat to the human population. Computer Vision, as a subfield of Artificial Intelligence, has enjoyed recent success in solving various complex problems in health care and has the potential to contribute to the fight of controlling COVID-19. In response to this problem, our motivation is to use computer vision based deep learning algorithms to develop an automated solution to prevent the spread of pandemic. Deep learning methods are used to build powerful face recognition systems capable of identifying and verifying the features of the infected person with high accuracy. The more training of models will improve the recognition accuracy of systems like humans.
In (CN2020/111024954A), the aqueous media gold immunohistochemistry detector seems to have an excellent sensitivity, an improved accuracy rate, and is easy to use. It requires no special devices or employee loyalty and productivity to operate, that can also be used for initial evaluation in a variety of settings, including neighborhood’s, predominant clinics, airport terminals, border checks, and sometimes even spouses and children. The system (US2020/10689716B1), is used track the existence of SARS-CoV-2 and minimize its dissemination, it's critical to diagnose contamination as quick as practicable with such an accurate, definitive diagnosis that can be used not just in clinics as well as in remote parts without the use of medical instruments. In (US2020/10822379B1), Methods and technologies for associating a binders to a SARS-CoV-2 peptide are described in this publication.
The (Hu et al [2017], Journal: Journal of biomedical optics, vol. 22, no. 3, pp. 036006), existing research shows that human physiological state can be perceived through breathing, which means respiratory signals are vital signs that can reflect human health condition to a certain extent. Many clinical literatures suggests that abnormal respiratory symptoms may be important factors for diagnosis of some specific diseases. For many people, early mild respiratory symptoms are difficult to be recognized. Therefore, through the measurement of respiration condition, potential COVID-19 patients can be screened to some extent. This may play an auxiliary diagnostic role, thus helping to find potential patients as early as possible (Jiang et al [2020], ArXiv abs/2004.06912).

Summary of the Invention
The proposed invention is to screen the COVID-19 affected patients automatically using facial recognition and thermal scanning, also to collect their facial features and temperature for tracking. The CNN model using collected dataset to increase the accuracy in cloud system. Along with this, to developed an assistive web-based tool for public and healthcare authorities to screen and track the infected patients.
Detailed Description of the Invention
Coronavirus Disease 2019 (COVID-19) has become a serious global epidemic in the past few months and caused huge loss to human society worldwide. For such a large-scale epidemic, early detection and isolation of potential virus carriers is essential to curb the spread of the epidemic. We propose an automated non-contact method to screen the health condition of people through facial analysis and thermal scanning. During the epidemic, many people tend to wear masks to reduce the risk of getting sick. The Multitask CNN based facial recognition algorithm provides better recognition under mask wearing conditions also. This may help identify those potential patients of COVID-19 under practical scenarios in various public localities. In this work, we will perform the health screening through the combination of the RGB and thermal scanning videos obtained from the dual-mode camera and deep learning architecture. We first accomplish a respiratory data capture technique for people wearing masks by using face recognition. Then, a bidirectional Multitask convolutional neural network with attention mechanism is applied to the respiratory data to obtain the health screening result.
Multi-task learning has been used successfully across all applications of machine learning, from natural language processing and speech recognition to computer vision and drug discovery. Given an image, we initially resize it to different scales to build an image pyramid, which is the input to the following three-step cascaded framework. We apply a Deep Multitask 3 Stage CNN neural network to do the classification task on judging whether the respiration condition is healthy or not. The input of the network is the respiration data obtained by our extraction method. Since the respiratory data is time series, it can be regarded as a time series classification problem. Therefore, we choose the Multitask Proposal and refine network model network with bidirectional and attention layers to work on the sequence prediction task. Among all the deep learning structures, MT-CNN is a type of neural network which is specially used to process time series data samples. After getting the masked region in thermal videos, the region of interest (ROI) that represents breathing features is obtained. When people wear masks, the nostrils are also blocked by the masks, and when people wearing different masks, the ROI may be different. Therefore, we perform a ROI tracking method based on maximizing the variance of thermal image sequence to extract a certain area on the masked region of the thermal video which stands for the breath signals most. Due to the lack of texture features in masked regions compared to human faces, we judge the ROI by the temperature change of thermal image sequence. The main idea is to traverse the masked region in the thermal images and find a small block with the largest temperature change as the selected ROI.
5 Claims & 1 Figure