Retracted: An automated face mask detection using machine learning techniques

In 2020, COVID has grown to be a significant public health concern. The virus's contact-transparent properties caused it to propagate quickly. Wearing a mask that stops the transmission is the most effective strategy to prevent this. The first step in lowering the danger is to wear a mask, which cuts the likelihood of contracting the infection by 70%. To resolve this problem, we need the system to keep an eye on the facial mask. This work can be facilitated by a face mask detector based on deep learning (DL) and machine learning (ML) models can make this task easier. Here, we do face mask categorization utilizing several ML and DL techniques. Here, we experimented with various methods such as CNN, Resnet50, Gabor filter, GLCM, SVM, KNN for extracting features and classifying faces with mask and without mask. From the accuracy we got for each model it says CNN performs well on the dataset we took.

Since the outbreak of the COVID-19 pandemic, the driven of face masks as personal protective equipment has increased significantly. Thus, disposed face masks from users should be handled properly for preventing the contamination of medical waste and the potential spread of viruses to the environment. This study gives information for dealing with face masks and assessing the volatile organic compounds (VOCs) concentrations via an atmospheric-pressure microwave plasma reactor. Face mask samples were mixed with the flux agents, namely cullet/glass, Al2O3, SiO2 and CaCO3. Samples were compared with control (no addition of flux agents) and the addition of only cullet. Moreover, microwave power, gas flow rate and pyrolysis duration were controlled at 1000 W, 9 standard liter per minute (SLM) and 5 min, respectively. The total concentration of VOC with the absence of flux agents was 448.04 ppm. Furthermore, the fuse of cullet and SiO2-Al2O3-cullet in the mask reduced the concentration of VOCs by 314.77 ppm and 54.7 ppm, respectively. Furthermore, the combination of CaCO3-SiO2-Al2O3-cullet creates the vitrification of material with the presence of crystalline structure, where the compositions of Ca and Si were 13.55% and 19.12%, respectively. Moreover, the final composition of carbon from the flux agents was 17.92 ± 10.08%. This study confirms that the fuse of CaCO3-SiO2-Al2O3-glass/cullet reduced the VOC via plasma technology, which is a promising method to be implemented in order to reduce the concentration of VOC from the face mask waste.

The spread of the coronavirus has been the focus the world's attention. The government has issued a new regulation on the coronavirus that obligates the entire community to always wear a face mask and implement social distancing to prevent the spread of viruses. These issues have been impacted by an increase in the use of face masks throughout the world and in Indonesia. Moreover, the World Health Organization (WHO) suggested using medical face masks. Furthermore, in the health care sector, the possible loading of pathogens in sub-micrometer sizes and properties such as splash resistance and the ability to prevent infection by reducing the concentration of inhaled particulates has created limitations on the types of face masks that can be used. Therefore it is necessary to design a face mask sterilizer device that can extend the life span of masks, thereby decreasing the masks consumption. An investigation on mask sterilizer devices based on heat pipes and thermoelectrics was conducted in this study. The objective of this study was to investigate the thermal performance of the mask sterilizer device. The method used was an experimental method using variations in the input voltage on the thermoelectric, namely 9 V, 10 V, and 11 V, and the microcontroller was also adjusted to control the temperature by 70 ℃, 80 ℃, and 90 ℃. The results showed that the thermoelectrics can generate heat with a temperature difference between the hot side and the cold side that can reach 80.11 ℃. In terms of voltage, the greater voltage that was given led to a greater resulting temperature of the sterilization device and reduced the amount of time to achieve the minimum sterilization temperature.KeywordsSterilizationFace maskHeat pipeThermoelectric

Facemask wearing is becoming a norm in our daily lives to curb the spread of Covid-19. Ensuring facemasks are worn correctly is a topic of concern worldwide. It could go beyond manual human control and enforcement, leading to the spread of this deadly virus and many cases globally. The main aim of wearing a facemask is to curtail the spread of the covid-19 virus, but the biggest concern of most deep learning research is about who is wearing the mask or not, and not who is incorrectly wearing the facemask while the main objective of mask wearing is to prevent the spread of the covid-19 virus. This paper compares three state-of-the- art object detection approaches: Haarcascade, Multi-task Cascaded Convolutional Networks (MTCNN), and You Only Look Once version 4 (YOLOv4) to classify who is wearing a mask, who is not wearing a mask, and most importantly, who is incorrectly wearing the mask in a real-time video stream using FPS as a benchmark to select the best model. Yolov4 got about 40 Frame Per Seconds (FPS), outperforming Haarcascade with 16 and MTCNN with 1.4. YOLOv4 was later used to compare the two datasets using Intersection over Union (IoU) and mean Average Precision (mAP) as a comparative measure; dataset2 (balanced dataset) performed better than dataset1 (unbalanced dataset). Yolov4 model on dataset2 mapped and detected images of masks worn incorrectly with one correct class label rather than giving them two label classes with uncertainty in dataset1, this work shows the advantage of having a balanced dataset for accuracy. This work would help decrease human interference in enforcing the COVID-19 face mask rules and create awareness for people who do not comply with the facemask policy of wearing it correctly. Hence, significantly reducing the spread of COVID-19.

COVID-19 pandemic has led to an international health emergency the WHO considers wearing a face mask an appropriate form of public health protection. This work will describe a face mask identification model that incorporates both deep and traditional machine learning techniques. Parts of the suggested model can be divided into two. Using Resnet50, the initial part of the system is set up for feature extraction. The second component classifies face masks using decision trees, support vector machines (SVMs), and the ensemble approach. The research will focus on three face-masked datasets. The Real-World Masked Face Dataset Includes three datasets: real-world masked faces, simulated faces, and wild faces (LFW). 99.64% of RMFD's SVM classifier is accurate throughout testing.. It achieved a 99.49% accuracy rate in SMFD and a 100% accuracy rate in LFW.

Face masks alter the speakers' voice, as their intrinsic properties provide them with acoustic absorption capabilities. Hence, face masks act as filters to the human voice. This work focuses on the automatic detection of face masks from speech signals, emphasising on a previous work claiming that face masks attenuate frequencies above 1 kHz. We compare a paralinguistics-based and a spectrograms-based approach for the task at hand. While the former extracts paralinguistic features from filtered versions of the original speech samples, the latter exploits the spectrogram representations of the speech samples containing specific ranges of frequencies. The machine learning techniques investigated for the paralinguistics-based approach include Support Vector Machines (SVM), and a Multi-Layer Perceptron (MLP). For the spectrograms-based approach, we use a Convolutional Neural Network (CNN). Our experiments are conducted on the Mask Augsburg Speech Corpus (MASC), released for the Interspeech 2020 Computational Paralinguistics Challenge (COMPARE). The best performances on the test set from the paralinguistic analysis are obtained using the high-pass filtered versions of the original speech samples. Nonetheless, the highest Unweighted Average Recall (UAR) on the test set is obtained when exploiting the spectrograms with frequency content below 1 kHz.

The new coronavirus, initially detected in Wuhan, China, has spread worldwide, wreaking devastation. The speed with which it spreads and the severity of the epidemic have made this a global emergency. COVID-19 is a very concerning disease from a public health perspective, making it critical to take precautions against its transmission, such as limiting close personal contact and using protective gear. The primary objective of this research is to create a system that can recognise human face masks and determine whether individuals are attempting to maintain social distance. Alterations to everyone's daily routines. In such stages, everyone must always keep their identity concealed. Because the massive number of populations has changed since the Outbreak of the Coronavirus pandemic, finding people who are not wearing veils is a challenge. The global spread of COVID-19 has altered society. Many of us are staying in our homes, avoiding contact with city dwellers, and adjusting our routines-such as when and where we go to work or school-in ways we never would have imagined. We need updated timetables as we transition away from outmoded procedures. What stands out the most is the widespread practise of hiding one's face behind a veil or other kind of covering whenever we enter a public building. Wearing a veil or covering one's face may provide some comfort while also preventing the spread of the COVID-19 virus. By preventing anybody, even the unwitting carriers, from spreading the virus, widespread usage of protective clothing has the potential to significantly reduce the rate of disease spread in a given area. Thus, the importance of the veil and its identification are made very plain. There has been a rise in the importance of face recognition frameworks, which are especially useful in hospitals and medical clinics where privacy of patients is a concern. They're also vital in places like airports, sports stadiums, warehouses, and other such crowded areas where heavy foot traffic necessitates strict security measures to ensure everyone's safety. The framework of face veil recognition can ensure our safety and the safety of those around us. This assignment may serve as a digitally administered test anywhere from a classroom to a hospital to a bank to an airport terminal. Through the use of photo processing and extensive learning, we are able to recognise human faces and separate them into two groups, those with and without head coverings. The assignment will let a person who is responsible for screening people to do so even if they are located at a remote location, while still being able to effectively screen and provide guidance. Open CV, Tensor Flow, and Keras are some of the Python libraries used. With Deep Learning, As part of their model preparation, these activities make use of Convolution Neural Networks, a subset of Deep Neural Networks.

This research paper inquiries about different attribute for purchasing face mask of Bangladeshi customer’s during pandemic circumstance. Specially, it reveals adequate information on the customer’s decision-making behavior and criteria for choosing toward buying face mask cloth. For the sake of this survey a questionnaire was employed as the tool to gather raw data, where 600 consumers were willingly attended from different region of three popular districts in Bangladesh. After collecting all raw data from respondents, the data of questionnaires was coordinated by Excel program, consequently the data was analyzed and integrated statistically to identify the customers attribute for satisfaction to face masks cloth. This process also detects numerous challenges and suggesting a proper solution to the problems that the textile industry facing in current situation. This research paper illustrates adequate information about the customer’s psychology in purchasing face masks cloth during pandemic situation which is available in the Bangladeshi market to ensure public health.

As per the reports of WHO, COVID-19 infection is generating a worldwide health catastrophe; hence the most practical prevention strategy is to use a face shield in public places. This epidemic forced administrative authorities all-round the globe to declare shutdowns in a mission to limit viral spread. According to research, using face coverings at work minimizes the danger of viral spread. An efficient and cost-effective method of applying Artificial Intelligence to maintain a protected workplace in industrial settings is developed. Regarding face mask identification, a hybrid approach based on deep learning techniques is employed. A facial mask identification database comprises photos with and without masks; OpenCV will be utilized to perform actual face detection via a live feed through the camera. Leveraging Python, and OpenCV, together with the DLIB algorithm, designers will create a face mask decoder with machine learning. The objective is to use machine learning algorithms and deep learning methods to determine whether the individual in the photo or footage is using a facial mask or not.

Since the infectious coronavirus disease (COVID-19) was first reported in Wuhan, it has become a public health problem around the world. This pandemic is having devastating effects on societies and economies. Due to the lack of health resources in a short period, all countries and continents are likely to face particularly severe damage that could lead to a large epidemic. Wearing a face mask that stops the transmission of droplets in the air can still be helpful in combating this pandemic. Therefore, this chapter focuses on implementing a face mask detection model as an embedded vision system. The six pre-trained models, which are MobileNet, ResNet-50, MobileNet-V2, VGG-19, VGG-16, and DenseNet, are used in this context. People wearing or not wearing masks were detected. After implementing and deploying the models, the selected models achieved a confidence score. Therefore, this study concludes that wearing face masks helps reduce the virus spread and fight this pandemic.

Detecting faces is a prevalent and substantial technology in current ages. It became interesting with the use of diverse masks and facial variations. The proposed method concentrates on detecting the facial regions in the digital images from real world which contains noisy, occluded faces and finally classification of images. Multi-task cascaded convolutional neural network (MTCNN)—a hybrid model with deep learning and machine learning to facial region detection is proposed. MTCNN has been applied on face detection dataset with mask and without mask images to perform real-time face detection and to build a face mask detector with OpenCV, convolutional neural networks, TensorFlow and Keras. The proposed system can be used as an application in the recent COVID-19 pandemic situations for detecting a person wears mask or not in controlling the spread of COVID-19.

The number of Covid-19 cases has decreased, with vaccines and the implementation of health protocols. Currently, people are starting to carry out face-to-face activities such as in schools, offices, shopping centers or other public places. However, entering the new normal activities, new variants of Covid19 may be formed and spread, therefore people still have to implement and maintain the health protocols, including the use of face mask in the offices or other public places. This study design the smart room system equipped with face mask detection to ensure the implementation of health protocols and air quality smart system to monitor the CO2 levels, temperature, and humidity, as well as disinfecting the room regularly with UV-C. Face Mask Detection system is developed in Raspberry, where the images from camera are processed with machine learning using MobileNetv2 for image classification method with Python libraries TensorFlow and Keras. The implementation results show that face mask detection able to process up to five people in one frame, with a maximum distance of 220cm from the camera. Air quality monitoring system also works well in detecting and classifying air parameters with error reading is under 5% and displaying them on the website application with average delay is about 4 seconds.

During the pandemic, the provision of Personal Protective Equipment (PPE) (e.g., face masks) to specialized personnel and general population represented a critical point to prevent virus spread; in addition, a huge variety of new manufacturers and products entered the market, widening the time required to test and validate the equipment’s performances according to the current regulations. In this study, we employ a simple method recently developed in our laboratory, to discriminate the filtering capability of a set of KN95-certified face masks from different producers. The method is based on a methylene blue (MB) solution which is sprayed towards a pneumatic lung simulator, connected to a dummy head dressed in different types of masks. The amounts of MB droplets passing through the tested devices are collected by a cotton pad and analyzed through UV-Vis Diffuse Reflectance. In the framework of interdisciplinary collaborations between the University and the Hospital, additional characterization techniques were performed including light optical and scanning electron microscopy morphological characterization, pressure decay, and bacterial filtration efficiency (BFE). All masks were compared to a reference one, considered the gold standard for filtering performances.

In today's digital environment, it's reassuring to know that analysis and modelling have gone into improving security in information systems via better trust management of personnel. Security mechanisms like trust are used to deal with varying degrees of authorization within an organization. Building and testing different machine learning models of trust based on the behaviour of an organization's employee data set is the first stage in our trust study. In this work, we show trust modelling on security systems using various machine learning (ML) techniques such as Random Forest (RF), Decision Tree (RF), XG Boost (XGB) and Logistic Regression (LR). We perform the training and the testing of our ML models based on stochastic pattern recognition to classify the Trust of an employee into four classes namely, Trusted, Moderate, U ntrusted and Unexpected. Later a rigorous comparison of all these models is done based on a Model Error Rate (MER) of a recommended trust board.

Cloud computing solutions are being gradually adopted in organizations, and this has raised a lot of questions about the risks that come with using cloud-based services, especially misconfiguration errors in cloud-based systems. Considering these challenges, the current study aimed to assess the use of AI-based security systems to identify misconfiguration errors in cloud computing platforms. The study has focused on the perceived effectiveness of implementing AI security systems for behavioral analysis, and the implementation of traditional security systems has been assessed. The study utilized an exploratory qualitative review approach analyzing the recent studies and the recent literature available on recognized databases of interest with a particular focus on the prevalence of misconfiguration errors, its importance for early detection, and the use of AI methods. The study results showed that the most significant threat and the main reason for cloud insecurity and breaches lie in cloud misconfiguration. However, the analysis of the study also exposed that AI and machine learning techniques have a promising future and could be enacted as an accurate technique for the automation of the early misconfiguration detection model. Furthermore, the research has also pointed out that planning and deception-based approaches and human factors analysis can also be used in an attempt to strengthen cloud security against misconfiguration threats. In general, the study has postulated that the enhancement of optimum AI-based security systems is fundamental to optimizing the misconfiguration error detection systems and the security of data in cloud computing.