Face Mask Classification based on Transfer Learning and MobileNet V2

Face mask image recognition can detect and monitor whether people wear the mask. Currently, the mask recognition model research mainly focuses on different mask detection systems. However, these methods have limited working datasets, do not give safety alerts, and do not work appropriately on masks. This paper aims to use the face mask recognition detection model in public places to monitor the people who do not wear the mask or the wrong mask to reduce the spread of Covid-19. The mask detection model supports transfer learning and image classification. Specifically, the collected data are first collected and then divided into two parts: with_mask and without_mask. Then authors build, implement the model, and obtain accurate mask recognition models. This paper uses and size of images datasets tested respectively. The experimental results show that the effect of the image size of was relatively better, and the training accuracy of different MobileNetV2 models is about 95%. Our analysis demonstrates that MobileNetV2 can correctly classify Covid-19. © 2022 ACM.

Efficient Transfer Learning for COVID-19 Diagnosis using X-Ray and Computed Tomography Images

In this work, we aim to find an effective model to diagnose COVID-19 by using a Transfer Learning (TL) model. The purpose is to classify COVID-19 infected persons from chest X-Ray (XR) and Computed Tomography (CT) images. Several Transfer Learning models have been studied to find the most efficient and effective among them. The proposed approach is based on Tensorflow and the architecture uses the MobileNet_V2 model. The datasets that are used in this study are publicly available. In order to train and evaluate our proposed model, we collected the CT scans dataset of 8000 images with two classes of infected and normal lungs, and the XR dataset contains 616 images. Two experiments are conducted with samples of different sizes to evaluate the model using google colab. The results revealed that the performance of our model MobileNet_V2 is highest with validation accuracy for XR and CT scans images: Val_AccuracyXR =96.77% and Val_AccuracyCT =99.67%, and test time for XR and CT scans images: TXR =0.18s, tCT=0.03s respectively. © 2022 IEEE.

Detection of Sars-Cov-2 from human chest CT images in Multi-Convolutional Neural Network's environment

This article focuses on the detection of the Sars-Cov2 virus from a large-scale public human chest Computed Tomography (CT) scan image dataset using a customized convolutional neural network model and other convolutional neural network models such as VGG-16, VGG-19, ResNet 50, Inception v3, DenseNet, XceptionNet, and MobileNet v2. The proposed customized convolutional neural network architecture contains two convolutional layers, one max pooling layer, two convolutional layers, one max pooling layer, one flatten layer, two dense layers, and an activation layer. All the models are applied on a large-scale public human chest Computed Tomography (CT) scan image dataset. To measure the performance of the various convolutional neural network models, different parameters are used such as Accuracy, Error Rate, Precision, Recall, and F1 score. The proposed customized convolutional neural network architecture's Accuracy, Error Rate, Precision Rate, Recall, and F1 Score are 0.924, 0.076, 0.937, 0.921, and 0.926 respectively. In comparison with other existing convolutional neural network strategies, the performance of the proposed model is superior as far as comparative tables and graphs are concerned. The proposed customized convolutional neural network model may help researchers and medical professionals to create a full-fledged computer-based Sars-Cov-2 virus detection system in the near future. © 2023 IEEE.

Comparative Analysis of Various Models for Potato Leaf Disease Classification using Deep Learning

Potato is one among the most extensively consumed staple foods, ranking fourth on the global food pyramid. Moreover, because of the global coronavirus outbreak, global potato consumption is expanding dramatically. Potato diseases, on the other hand, are the primary cause of crop quality and quantity decline. Plant conditions will be dramatically worsened by incorrect disease classification and late identification. Fortunately, leaf conditions can help identify various illnesses in potato plants. Potato (Solanum tuberosum L) is one of the majorly farmed vegetable food crops in worldwide. The output of potato crops in both quality and quantity is affected majorly due to fungal blight infections, which causes a severe impact on the global food yield. The most severe foliar diseases for potato crops are early blight and late blight. The causes of these diseases are Alternaria solani and Phytophthora infestants respectively. Farmers suspect such problems by focusing on the color change or transformation in potato leaves, which is effortless due to subjectivity and lengthy time commitment. In such circumstances, it is critical to develop computer models that can diagnose those diseases quickly and accurately, even in their early stages. © 2023 IEEE.

Real-time Face mask Detection Using Deep Learning and MobileNet V2

COVID-19 has affected the whole world severely. Lockdowns and quarantines are imposed all over the world to prevent its spread. Hand sanitizers and face masks were made compulsory for individuals to apply for safety of their own and their society. This project will check the presence or the absence of masks on the face of a person. There could be more than a single person in the input provided, and the input could vary from images to GIFs to Livestreams. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Face Mask Detection and Recognition System

As today a disease called COVID-19 is causing health crisis and deaths, it became most essential to wear a mask for protecting ourselves from Corona virus. Even in public areas, where is more rush we should wear mask as no virus can spread from person to person if any one of person from public is Corona positive. This paper introduces face mask detection that can be used by the authorities to make mitigation, evaluation, prevention, and action planning against COVID19. So basically in this project we are going to use Python, Keras, OpenCV alongwith MobileNet for this Face Mask Detection System. This includes some steps like data preprocessing, training and testing the model, run and view the accuracy and applying model in the camera. The inputs has provided here are 1000+ images of people with mask and without mask. First the data get processed and then by checking features of each image it will train all the models and the persons with and without mask get separated to two categories: with mask and without mask. If person is wearing mask with 90 or more percent of accuracy, then he will get added to with mask category and person not wearing mask get added to without mask category, so that we can permit with mask person to public areas. © 2022 IEEE.

Detection of COVID-19 Infection in CT and X-ray images using transfer learning approach.

BACKGROUND: The infection caused by the SARS-CoV-2 (COVID-19) pandemic is a threat to human lives. An early and accurate diagnosis is necessary for treatment. OBJECTIVE: The study presents an efficient classification methodology for precise identification of infection caused by COVID-19 using CT and X-ray images. METHODS: The depthwise separable convolution-based model of MobileNet V2 was exploited for feature extraction. The features of infection were supplied to the SVM classifier for training which produced accurate classification results. RESULT: The accuracies for CT and X-ray images are 99.42% and 98.54% respectively. The MCC score was used to avoid any mislead caused by accuracy and F1 score as it is more mathematically balanced metric. The MCC scores obtained for CT and X-ray were 0.9852 and 0.9657, respectively. The Youden's index showed a significant improvement of more than 2% for both imaging techniques. CONCLUSION: The proposed transfer learning-based approach obtained the best results for all evaluation metrics and produced reliable results for the accurate identification of COVID-19 symptoms. This study can help in reducing the time in diagnosis of the infection.

Real-time Face mask Detection Using Deep Learning and MobileNet V2

COVID-19 has affected the whole world severely. Lockdowns and quarantines are imposed all over the world to prevent its spread. Hand sanitizers and face masks were made compulsory for individuals to apply for safety of their own and their society. This project will check the presence or the absence of masks on the face of a person. There could be more than a single person in the input provided, and the input could vary from images to GIFs to Livestreams. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Deep Neural Network models for diagnosis of COVID-19 Respiratory diseases by analyzing CT-Scans and Explain-ability using trained models

The Novel Coronavirus, popularly known as "COVID-19,"is causing a pandemic over the world. This virus causes severe respiratory disease in those who are afflicted. Symptoms such as fever, dry cough, and exhaustion can be used to identify this virus.These symptoms, on the other hand, are comparable to those of other viral or respiratory illnesses. There is no quick method to tell whether or not someone has been exposed to the virus.. To counter the aforementioned constraints, a quicker diagnosis is desired, which brings us to the study's goal: to develop a diagnostic approach that incorporates previous data, mostly from COVID-19, as well as data-sets from other respiratory disorders. Deep learning models will be used to evaluate the data sets we've gathered, helping us to make more accurate and efficient decisions. convolutional Neural Network models such as VGG 19, Inception v3, MobileNet V2, and ResNet 50 are among the Deep Neural Network models we plan to deploy. These four models have been pre-trained to categorize CT-Scan images using trained learning methodologies. To obtain faster and more accurate answers, the outcomes of each model are compared among the models. A "Hybrid"model built of Convolutional Neural Network and a Support Vector Machine is also proposed in this research. The Hybrid Model is not as deep as the pre-trained models, but it is as accurate. We will be able to diagnose more correctly and effectively based on the correctness of the outcome and the shortest time necessary for categorization of images which will enable us to diagnose more accurately and effectively.In our research work we have collected data-sets from git-hub [19] and Kaggle and in total we have gathered 877 images of chest X-rays and CT-scans. We operated data-augmentation and smote analysis on our data-set. After training and testing our models we have obtained the following accuracy scores: 0.9384 validation-accuracy and 0.9361 train-accuracy for Hybrid model, 0.9806 validation-accuracy and 0.9692 train-accuracy for Inception-V3, 0.9806 validation-accuracy and 0.9846 train-accuracy for MobileNet V2, 0.7206 validation-accuracy and 0.6705 train-accuracy for Resnet 50, 0.9107 validation-accuracy and 0.9685 train-accuracy for VGG 19. © 2022 IEEE.

Masked Face Recognition and Liveness Detection Using Deep Learning Technique

Face recognition has been the most successful image processing application in recent times. Most work involving image analysis uses face recognition to automate attendance management systems. Face recognition is an identification process to verify and authenticate the person using their facial features. In this study, an intelligent attendance management system is built to automate the process of attendance. Here, while entering, a person’s image will get captured. The model will detect the face;then the liveness model will verify whether there is any spoofing attack, then the masked detection model will check whether the person has worn the mask or not. In the end, face recognition will extract the facial features. If the person’s features match the database, their attendance will be marked. In the face of the COVID-19 pandemic, wearing a face mask is mandatory for safety measures. The current face recognition system is not able to extract the features properly. The Multi-task Cascaded Convolutional Networks (MTCNN) model detects the face in the proposed method. Then a classification model based on the architecture of MobileNet V2 is used for liveness and mask detection. Then the FaceNet model is used for extracting the facial features. In this study, two different models for the recognition have been built, one for people with masks another one for people without masks. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

AUTOMATIC FACE MASK DETECTION BASED ON MOBILENET V2 AND DENSENET 121 MODELS

The COVID-19 pandemic has brought significant impacts to the world. In Indonesia, public places such as malls, restaurants, shops, private and government offices, and public areas obliged visitors to wear masks. Unfortunately, there are times when visitors do not obey the rules by not wearing a mask;therefore, surveillance must be conducted. However, manual surveillance to check if a person wearing a mask can be a tedious task. This research aims to propose an automatic face mask detection that can detect if a person is using a mask or not. The proposed method combines face detection and classification using deep learning. The face detection is done using USM sharpening, CenterFace, and two pre-trained models, the MobileNet V2 and DenseNet 121 are used to classify if a person wears a face mask or not. The pre-trained models were fine-tuned using two datasets. Google Colab and libraries such as Tensorflow, Keras, and Scikit-learn were utilized. The research results show that the MobileNet V2 achieves higher performance and has a faster execution time. © 2022 ICIC International. All rights reserved.

Efficiency for Face Mask Detection in Neural Network

Because of COVID-19, wearing a face mask has become the most efficient and convenient way to spread this virus. Face mask detection can fulfill the function of warning those people who do not wear a face mask. Using the Convolutional neural network, the Feedforward Neural Network and the MobileNet V2, a high recognition rate for the face mask detecting system can be achieved. This study compares the accuracy, the loss and the training time for these models and concludes that CNN is the best model based on its high accuracy of 100%. The result that comes out from our study can improve the efficiency of the face mask detecting system. In general, the identification model in our study can be changed easily to apply in other areas, such as medical image classification and geographic image classification. © 2021 IEEE.

COVID-19 Outbreak : Detecting face mask types in real time

While the COVID-19 outbreak poses several major hazards to the world, it also serves as a reminder that we must take care to prevent the virus from spreading. Wearing a mask is one of the most effective non-pharmaceutical strategies for preventing the spread of infectious diseases. Therefore, to aid in the prevention of a public epidemic, an automatic real-time mask recognition and categorization solution is urgently required. The efficiency of facemasks has been called into question, owing to poor mask selection. N95 masks must be worn in jobs where there is a high danger of contracting the virus. Surgical, DIY, and N95 masks all have varied degrees of effectiveness. To safeguard public safety, we created a deep learning model that can classify different types of masks as well as determine if there is no mask in real-time video or images. Our framework consists of three main phases: Face detection using object detection API, application of face mask classifier and classification of masks into four different categories. In this study, we show that SSD MobileNet V2 outperforms both SSD MobileNet V1. and SSD Resnet50 V1. in terms of sensitivity and precision.

Automated Facemask Detection and Monitoring of Body Temperature using IoT Enabled Smart Door

Living with the novel Coronavirus is becoming the new normal as nations around the globe resume. However, in order to stop the virus from spreading, we must isolate Covid-infected persons from the rest of the population.Fever is the most common symptom of coronavirus infection, according to the CDC [1], with up to 83 percent of symptomatic patients presenting indications of fever. Early symptom detection and good hygiene standards are therefore critical, particularly in situations where people come into random contact with one another. As a result, temperature checks and masks are now required in schools, colleges, offices, and other public spaces. However, manually monitoring each individual and measuring their respective body temperatures is a cumbersome task. Currently, most of the temperature checkups are done manually which can be inefficient, impractical, and riskybecause sometimes people checking manually may be reluctant to check every person’s temperature or sometimes allow people even if they violate the guidelines. Moreover, the person assigned to manually check will be at high risk as he is exposed to a lot of people. To solve these issues, we propose a project that reduces the growth of COVID-19 by monitoring the presence of a facial mask and measuring their temperature. The Face Mask Detection can be done using the TensorFlow software library, Mobilenet V2 architecture and OpenCV.A non-contact IR temperature sensor is used to monitor the individual's body temperature. To avoid false positives, the system will be strengthened by training it with a variety of cases. Once the system detects a mask, it measures the body temperature of the person. If the temperature is within the normal range, sanitization is done,and the person is permitted entry through an IOT enabled smart door. However, if the system fails to detect a mask or the person's temperature falls out of the predefined range, a buzzer rings and the door remains closed. Our model is intended to be effective in preventing the spread of this infectious disease. © 2022 IEEE.

Deep Learning for Real-Time Face Mask Detection

The corona virus (COVID19) pandemic requires immediate action to avoid adverse effects on local health and the global economy. Due to the effects of COVID19, most of the people lives have been reversed. In the absence of effective antivirals and inadequate medical resources, UN agencies propose a number of measures to regulate infection rates and prevent the limited medical resources from being exhausted. Wearing a face mask is a type of the non-pharmaceutical intervention techniques that can block the primary care of viral droplets ejected by an infected person. According to government basics, it is important for everyone in every country to wear a mask. The government recommends wearing a mask, but many do not. Mask detection is very important in this situation. To contribute to community health, this study aims to develop highly accurate and timely techniques for detecting non-face masks in public and encouraging people to use them. It is said that "Increase the number of people who wear masks correctly and reduce the number of infected people". Starting with MobileNet V2 as a baseline, we used the concept of transfer learning to fuse high levels of linguistic data during mask recognition. For the face detection module, we used Caffe Model in conjunction with OpenCV's DNN module. The anticipated model's remarkable performance makes it ideal for live video police work equipment that detects face masks in real - time. © 2022 IEEE.

Deep Learning Based Intelligent Classification of Covid-19 Pneumonia Using Cough Auscultations

The World Health Organization has designated COVID-19 a pandemic because its emergence has influenced more than 50 million world's population. Around 14 million deaths have been reported worldwide from COVID-19. In this research work, we have presented a method for autonomous screening of COVID-19 and Pneumonia subjects from cough auscultation analysis. Deep learning-based model (MobileNet v2) is used to analyze a 6757 self-collected cough dataset. The experimentation has demonstrated the efficiency of the proposed technique in distinguishing between COVID-19 and Pneumonia. The results have demonstrated the cumulative accuracy of 99.98%, learning rate of 0.0005 and validation loss of 0.0028. Furthermore, cough analysis can be performed for other patients screening of other pulmonary abnormalities. © 2021 IEEE.

A Deep Comprehensive 3D Modelling for the Prediction of Coronavirus on Medical Scans of COVID-19 Patients

The global pandemic of coronavirus disease 19 (COVID-19) left many lives in question with the virus, and many people are being suffered every day. In this hectic situation, each patient needs to get attended by the doctor for the scans and determine if they are positive with the virus to move on to the diagnosis stage. It is hard to attend each patient when many people are being tested and effected. It is time where machine learning and artificial intelligence are used. A lot of efforts were made by the researchers to diagnose the effected patients;using 2D images in present, we present an observation that helps detect COVID-19 with the use of 3D images. In this paper, we have worked with 3D image data and have found one of the best models among Inception v3, ResNet-50, VGG16, EfficientNet B0, MobileNet v2. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Social Distancing and Face Mask Monitoring System Using Deep Learning Based on COVID-19 Directive Measures

We, the entire world is in the lock of a micro size virus named Corona we are in the urge of saving our life rather than the money. This virus had changed the attitude of people from generations together, in this two years people realized that their health worth more than their net worth. We are in an uncertain situation but, we can bring the world back to normal so, we need to follow the guidelines issued by the health organizations so our government insisted people wear the mask and maintain social distance to control the spread of the disease but 90% percent of people not following covid guidelines. The main motive in this paper, mask detection on face with social distancing which helps to overcome this pandemic situation. Our proposed system comprises of data processing, data augmentation, image classification using mobilenetv2 and object detection plays a vital role in this paper. The modules are developed using TensorFlow and open-cv python programming to detect the faces with mask. If a person wears a mask they will be in a safe zone and the system shows a green box where if the person doesn't wear a mask, then it will be shown in a red box and with the message of alert as well. Social distancing detection will detect that two or more person in a single frame are walking with maintaining social distancing with at least 2 meters of range with each other using the Euclidean distance method, it will work in a Reliable manner with accurate results during this current situation which will easily help to track the person and collect fine if they violate any government directive guidelines so our system, will prevent the spread of the disease. Every Automation process reduces manual inspection to inspect the people which can be used in public places to control the spread of the virus and this prototype could be used in many places like park, hospital, airports, temples, railway station etc.To control this pandemic situation © 2021 IEEE.

Deep learning empowered COVID-19 diagnosis using chest CT scan images for collaborative edge-cloud computing platform.

The novel coronavirus outbreak has spread worldwide, causing respiratory infections in humans, leading to a huge global pandemic COVID-19. According to World Health Organization, the only way to curb this spread is by increasing the testing and isolating the infected. Meanwhile, the clinical testing currently being followed is not easily accessible and requires much time to give the results. In this scenario, remote diagnostic systems could become a handy solution. Some existing studies leverage the deep learning approach to provide an effective alternative to clinical diagnostic techniques. However, it is difficult to use such complex networks in resource constraint environments. To address this problem, we developed a fine-tuned deep learning model inspired by the architecture of the MobileNet V2 model. Moreover, the developed model is further optimized in terms of its size and complexity to make it compatible with mobile and edge devices. The results of extensive experimentation performed on a real-world dataset consisting of 2482 chest Computerized Tomography scan images strongly suggest the superiority of the developed fine-tuned deep learning model in terms of high accuracy and faster diagnosis time. The proposed model achieved a classification accuracy of 96.40%, with approximately ten times shorter response time than prevailing deep learning models. Further, McNemar's statistical test results also prove the efficacy of the proposed model.