Building a Real-Time System to Monitor Students Electronically Based on Digital Images of Face Movement

The spread of the Corona Virus pandemic on a global scale had a great impact on the trend towards e-learning. In the virtual exams the student can take his exams online without any papers, in addition to the correction and electronic monitoring of the exams. Tests are supervised and controlled by a camera and proven cheat-checking tools. This technology has opened the doors of academic institutions for distance learning to be wide spread without any problems at all. In this paper, a proposed model was built by linking a computer network using a server/client model because it is a system that distributes tasks between the two. The main computer that acts as a server (exam observer) is connected to a group of sub-computers (students) who are being tested and these devices are considered the set of clients. The proposed student face recognition system is run on each computer (client) in order to identify and verify the identity of the student. When another face is detected, the program sends a warning signal to the server. Thus, the concerned student is alerted. This mechanism helps examinees reduce cheating cases in early time. The results obtained from the face recognition showed high accuracy despite the large number of students' faces. The performance speed was in line with the test performance requirements, handling 1,081 real photos and adding 960 photos. © 2022 IEEE.

Design of an Intelligent Educational Evaluation System Using Deep Learning

Nowadays, online education has been a more general demand in context of COVID-19 epidemic. The intelligent educational evaluation systems assisted by intelligent techniques are in urgent demand. To deal with this issue, this paper introduces the strong information processing ability of deep learning, and proposes the design of an intelligent educational evaluation system using deep learning. Inside the algorithm part, the low-complexity offset minimal sum (OMS) is selected as the front-end processor of deep neural network, so as to reduce following computational complexity in deep neural network. And the deep neural network is adopted as the major calculation backbone. In this paper, our OMS deep neural network parameters are 23 and 57 compared with other parameters, which can save about 59.64% of the network parameters, and the training time is 11270 s and 25000 s respectively, which saves the training time 54.92%. It can be also reflected from experiments that the proposal further improves the performance of unbalanced data classification in this problem scenario. © 2013 IEEE.

Real-time Control of the Air Volume in Ventilation Facilities by Limiting Concentration with Cluster Algorithms

Ventilation improves indoor air quality and reduces airborne infections. It is particularly important at present because of the COVID-19 pandemic. Commercially available ventilation facilities can only be instantly turned on/off or at a set time with adjustable air volumes (high, middle, and low). However, maintaining the indoor carbon dioxide concentration while reducing the energy consumption of these facilities is challenging. Hence, this study developed clustering algorithms to determine the carbon dioxide concentration limit thus enabling real-time air volume adjustment. These limit values were set using the existing energy recovery ventilation (ERV) controller. In the experiment, dual estimation was adopted, and the constructing building energy models from data were sampled at a low rate to compare that the ventilation facilities are only turned on/off. In addition, switching control is closely related to fuzzy control;that is, fuzzy control can be considered a smooth version of switching control. The experimental results indicated that the limits of 600 and 700 ppm were suitable to effectively control the real-time air volume based on the ERV operation. An ERV-based carbon dioxide concentration limit reduced the energy consumption of ventilation facilities by 11%implications of this study. IEEE

The IMU Augment SLAM on Unmanned Vehicle for Detection of Protective Measures in COVID-19

Detecting protective measures (e.g., masks, goggles and protective clothing) is a momentous step in the fight against COVID-19. The detection mode of unmanned devices based on Simultaneous localization and mapping (SLAM) and fusion technology is more efficient, economical and safe than the traditional manual detection. In this paper, a tightly-coupled nonlinear optimization approach is used to augment the visual feature extraction of SLAM by the gyroscope of the IMU to obtain a high-precision visual inertial system for joint position and pose estimation. Based on the VINS-Mono frame, first, an LSD algorithm based on a conditional selection strategy is proposed to extract line features efficiently. Then, we propose recovering missing point features from line features. Moreover, we propose a strategy to recover vanishing point features from line features, and add residuals to the SLAM cost function based on optimization, which optimizes point-line features in real time to promote the tracking and matching accuracy. Second, the wavelet threshold denoising method based on the 3σ criterion is used to carry out real-time online denoising for gyroscope to improve the output precision. Our WD-PL-VINS was measured on publicly available EuRoC datasets, TUM VI datasets and evaluated and validated in lab testing with a unmanned vehicle (UV) based on the NVIDIA Jetson-TX2 development board. The results show that our method's APE and RPE on MH-03-easy sequences are improved by 69.28% and 97.66%, respectively, compared with VINS-Mono. © 2001-2012 IEEE.

Two-Stage OD Flow Prediction for Emergency in Urban Rail Transit

Urban rail transit (URT) is vulnerable to natural disasters and social emergencies including fire, storm and epidemic (such as COVID-19), and real-time origin-destination (OD) flow prediction provides URT operators with important information to ensure the safety of URT system. However, hindered by the high dimensionality of OD flow and the lack of supportive information reflecting the real-time passenger flow changes, study in this area is at the beginning stage. A novel model consisting of two stages is proposed for OD flow prediction. The first stage predicts the inflows of all stations by Long Short-Term Memory (LSTM) in real time, where the dimension is reduced compared with predicting OD flows directly. In the second stage, the notion of separation rate, namely, the proportion of inbound passengers bounding for another station, is estimated. Finally, The OD flow is predicted by multiplying the inflow and separation rate. Experiments based on Hangzhou Metro dataset show the proposed model outperforms the contrast model in weighted mean average error (WMAE) and weighted mean square error (WMSE). Results also suggest that the proposed prediction model performs better on weekdays than on weekends, and with greater accuracy on larger OD flows. IEEE

Facial mask detection system based on YOLOv4 algorithm

This paper mainly addresses the detection of facial mask wear under the new COVID-19. To meet this demand, this paper performs facial mask wear detection on specific targets through a model trained based on the YOLOv4 algorithm. It has the characteristics of fast detection and light weight, and the application of this system to daily mask wear detection requires high real-time system performance. YOLOv4 meets this requirement, so the system designed based on this model has practical significance. This paper further demonstrates that the facial mask detection system designed based on the YOLOv4 algorithm is capable of working in multiple scenes of daily life, successfully detecting whether the target is wearing a mask in many scenes such as routine, multi-person and occlusion environment. © 2022 IEEE.

Game Theory-inspired Novel Mechanism for Assessing Healthcare Quality

Healthcare is the most pivotal domain of every nation. With the sudden upraise of the COVID-19 pandemic, there has been a major concern for the healthcare industry to provide quality medical services to the common people. Vulnerable healthcare conditions have proven to be fatal for the patients. Conspicuously, it has become indispensable to assess the quality of healthcare services provided by the hospitals. The current paper focuses on analyzing healthcare service quality delivered by the hospitals and healthcare centers. Specifically, the presented framework utilizes Internet of Things (IoT) technology to acquire real-time ambient data inside smart hospitals. The quantification of the healthcare service is performed using Probability of Health Grade (PoHG) to classify data segments using the Probabilistic Bayesian Belief Model. Furthermore, the temporal data ion is performed for the numerical analysis of healthcare service quality in terms of the Health Quality Index (HQI). Finally, a 2-player game theory-inspired decision modeling is performed to analyze healthcare quality in a time-sensitive manner. The proposed framework is assessed using a simulated environment where 225,325 data segments are analyzed. Results are compared with state-of-the-art techniques in which enhanced performance measures are registered in terms of Classification Efficacy (93.74%), Decision-Making Efficiency (Coefficient of Determination (95%), Accuracy (97.53%), Mean Square Error(2.01%)), Root Mean Square Error(1.95%)), Temporal Delay (96.62s), and Reliability(91.58%). IEEE

The IMU augment SLAM on Unmanned Vehicle for detection of protective measures in COVID-19

Detecting protective measures (e.g., masks, goggles and protective clothing) is a momentous step in the fight against COVID-19. The detection mode of unmanned devices based on Simultaneous localization and mapping (SLAM) and fusion technology is more efficient, economical and safe than the traditional manual detection. In this paper, a tightly-coupled nonlinear optimization approach is used to augment the visual feature extraction of SLAM by the gyroscope of the IMU to obtain a high-precision visual inertial system for joint position and pose estimation. Based on the VINS-Mono frame, first, an LSD algorithm based on a conditional selection strategy is proposed to extract line features efficiently. Then, we propose recovering missing point features from line features. Moreover, we propose a strategy to recover vanishing point features from line features, and add residuals to the SLAM cost function based on optimization, which optimizes point-line features in real time to promote the tracking and matching accuracy. Second, the wavelet threshold denoising method based on the 3σcriterion is used to carry out real-time online denoising for gyroscope to improve the output precision. Our WD-PL-VINS was measured on publicly available EuRoC datasets, TUM VI datasets and evaluated and validated in lab testing with a unmanned vehicle (UV) based on the NVIDIA Jetson-TX2 development board. The results show that our method’s APE and RPE on MH 03 easy sequences are improved by 69.28% and 97.66%, respectively, compared with VINS-Mono. IEEE

Application of Deep Learning in Detection of Covid-19 Face Mask

With the increase in the number of Covid-19 cases throughout the globe wearing face masks has proved to be effective in the prevention of the virus. In this work, we have originated a method that can detect if people are violating the rule of wearing a mask outdoors using a two-stage deep learning system. The first stage of the system detects different faces present in the input image using YOLO (You Only Look Once) model trained for the face detection and returns face ROIs. In the second stage extracted face ROI is passed through face mask detector model trained using MobileNetV2 which in turn classifies it as Mask or No mask. The dataset used for training the mask detector model is Real-World Masked Face Dataset (RMFD) and for Face Detection model is the WIDER dataset. The proposed method gives 98% accuracy for mask detection. The promising results derived from the proposed model demonstrate that the deployment of the model can be done in real-time systems. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

IoT Enabled Smart Aiport System to Avoid Covid-19

Today's generation wants everything easier, faster and automatic. During this corona pandemic, health and safety of each and every individual, either who is traveling through flights or working at an airport is a big issue. Usually, when we go to an airport, we go through many checks, and before boarding the flight, the security check-in, our luggage bags are counted and tagged by the person working at the counter of the airport. The luggage bags are put on the conveyor belt and the person working at the counter has to count the luggage bags by himself, he has to stick the tags on the luggage bags. None of the airports provides the facility of automatic counting of the luggage bags and sticking tags on them. And during this COVID-19 pandemic, we should avoid touching maximum things. This research paper provides a new technique for the same and that in a smart way. In this research, we are providing a novel approach to create an automatic system which will help to make the airport a smart one with IOT sensors and devices. Smart Airport also provides the counting of the luggage bags, tagging of the luggage bags, checking the presence of metallic objects in the luggage bags in a single embedded system. This approach will help the human society in maintaining social distancing and help them to save their time. © 2021 IEEE.