Convolutional Neural Network-based Architecture for Detecting Face Mask in Crowded Areas DOI
Jad Abou Chaaya,

Batoul Zaraket,

Hassan Harb

et al.

Published: July 2, 2023

After the invasion of Covid-19 virus, governments started containing spread virus by forcing people to wear face masks in public places. Therefore, automatic mask detection has become very important limit spread. Unfortunately, existing methods present limited performance accurately detecting crowded areas due significant number faces per scene. In order tackle this challenge, we propose a two-stage neural network-based architecture that can detect environments. Several simulations have been conducted investigate efficiency proposed and results show high accuracy reach up 96.5%.

Language: Английский

CNN BASED PATHWAY CONTROL TO PREVENT COVID SPREAD USING FACE MASK AND BODY TEMPERATURE DETECTION DOI Open Access

T Gowtham Prasad,

Anil V. Turukmane, M. Sunil Kumar

et al.

Journal of Pharmaceutical Negative Results, Journal Year: 2022, Volume and Issue: 13(SO4)

Published: Jan. 1, 2022

Airborne diseases cause detriment in the human life.Many were found history such as TB, SARS, MERS and recently COVID 19.These hit dead rate crushes health wealth of world population.Mostly, airborne will spread rapidly crowdy places.Especially, case COVID-19, Wearing mask monitoring body temperature by individual is good solution to prevent rapid disease.So, keeping safety measures face crucial places Airports, railway stations, Bus Stations, malls, temples, etc. obligatory.With a focus on emphasizing people we proposed integrated system that monitors each open/close pathway gate allow after knee verification.Proposed Prototype uses Raspberry pi monitor Face using CNN Arduino enable motor drivers open or close Gate.Efficiency loss Proposed model was trained tested with multiple epoches.

Language: Английский

Citations

8
Wearing face mask detection using deep learning through COVID-19 pandemic DOI Open Access

Javad Khoramdel,

Soheila Hatami,

Majid Sadedel

et al.

Scientia Iranica, Journal Year: 2023, Volume and Issue: 0(0), P. 0 - 0

Published: Feb. 13, 2023

During the COVID-19 pandemic, wearing a face mask has been known to be an effective way prevent spread of COVID-19. In lots monitoring tasks, humans have replaced with computers thanks outstanding performance deep learning models. Monitoring is another task that can done by models acceptable accuracy. The main challenge this limited amount data because quarantine. paper, we did investigation on capability three state-of-the-art object detection neural networks for real-time applications. As mentioned, here are used, Single Shot Detector (SSD), two versions You Only Look Once (YOLO) i.e., YOLOv4-tiny, and YOLOv4-tiny-3l from which best was selected. proposed method, according different models, model suitable use in real-world mobile device applications comparison other recent studies YOLOv4-tiny model, 85.31% 50.66 mean Average Precision (mAP) Frames Per Second (FPS), respectively. These values were achieved using datasets only 1531 images separate classes.

Language: Английский

Citations

4
Cell recognition based on atomic force microscopy and modified residual neural network DOI

Junxi Wang,

Mingyan Gao,

Lixin Yang

et al.

Journal of Structural Biology, Journal Year: 2023, Volume and Issue: 215(3), P. 107991 - 107991

Published: July 13, 2023

Language: Английский

Citations

4
MobileNet-Powered Deep Learning for Efficient Face Classification DOI

Pavani Chitrapu,

Hemantha Kumar Kalluri

2020 IEEE International Students' Conference on Electrical,Electronics and Computer Science (SCEECS), Journal Year: 2024, Volume and Issue: unknown, P. 1 - 6

Published: Feb. 24, 2024

Facial image analysis and categorization have recently made great strides in computer vision. The current study, explores ways to help computers better recognize faces quickly accurately, especially for tasks like security entertainment. Identifying faces, emotions, identities is crucial Security Surveillance, Access Control, user Authentication Smart Devices, Emotion Analysis Human-Computer Interaction. Adopting the MobileNet deep learning model because it requires less memory works efficiently. To make even more effective at recognizing adjusted its parameters tested with two data sets, CASIA 3D face set 105 pins set. study using MobileNetV2 achieved a very high accuracy of 98.71% on 99.29% experimental results show that understands different situations.

Language: Английский

Citations

1
Utilization of convolutional neural networks to analyze microscopic images for high-throughput screening of mesenchymal stem cells DOI Creative Commons

MuYun Liu,

Xiangxi Du, Junyuan Hu

et al.

Open Life Sciences, Journal Year: 2024, Volume and Issue: 19(1)

Published: Jan. 1, 2024

Abstract This work investigated the high-throughput classification performance of microscopic images mesenchymal stem cells (MSCs) using a hyperspectral imaging-based separable convolutional neural network (CNN) (H-SCNN) model. Human bone marrow (hBMSCs) were cultured, and acquired fully automated microscope. Flow cytometry (FCT) was employed for functional classification. Subsequently, H-SCNN model established. The (HSM) created, spatial-spectral combined distance (SSCD) to derive neighbors (SSNs) each pixel in training set determine optimal parameters. Then, CNN (SCNN) adopted instead classic layer. Additionally, cultured seeded into 96-well plates, high-functioning hBMSCs screened both manual visual inspection (MV group) (H-SCNN group), with group consisting 96 samples. FCT served as benchmark compare area under curve (AUC), F 1 score, accuracy (Acc), sensitivity (Sen), specificity (Spe), positive predictive value (PPV), negative (NPV) between groups. best Acc 0.862 when window size 9 12 SSNs. SCNN model, ResNet VGGNet gradually increased increase sample size, reaching 89.56 ± 3.09, 80.61 2.83, 80.06 3.01%, respectively at 100. corresponding time significantly shorter 21.32 1.09 min compared (36.09 3.11 min) models (34.73 3.72 ( P < 0.05). Furthermore, AUC, Acc, Sen, Spe, PPV, NPV all higher group, less required Microscopic based on proved be effective assessment hBMSCs, demonstrating excellent efficiency, enabling its potential powerful tool future MSCs research.

Language: Английский

Citations

1
Deep learning based masked face recognition in the era of the COVID-19 pandemic DOI Open Access
Ashwan A. Abdulmunem,

Noor D. Al-Shakarchy,

Mais Saad Safoq

et al.

International Journal of Power Electronics and Drive Systems/International Journal of Electrical and Computer Engineering, Journal Year: 2022, Volume and Issue: 13(2), P. 1550 - 1550

Published: Dec. 11, 2022

<span lang="EN-US">During the coronavirus disease 2019 (COVID-19) pandemic, monitoring for wearing masks obtains a crucial attention due to effect of prevent spread coronavirus. This work introduces two deep learning models, former based on pre-trained convolutional neural network (CNN) which called MobileNetv2, and latter is new CNN architecture. These models have been used detect masked face with three classes (correct, not correct, no mask). The experiments conducted benchmark dataset mask detection from Kaggle. Moreover, comparison between driven evaluate results these proposed models.</span>

Language: Английский

Citations

6
Neural networks contribution in face mask detection to reduce the spread of COVID-19 DOI Open Access
Maminiaina Alphonse Rafidison, Andry Harivony Rakotomihamina, Sabine Harisoa Jacques Rafanantenana

et al.

Multimedia Tools and Applications, Journal Year: 2023, Volume and Issue: 82(21), P. 32559 - 32581

Published: March 4, 2023

Language: Английский

Citations

3
Auto Detection of Personal Protective Equipment on Human DOI
Chandu Vaidya,

Prajakta Yelure,

Shivani Gonnade

et al.

2020 IEEE International Students' Conference on Electrical,Electronics and Computer Science (SCEECS), Journal Year: 2023, Volume and Issue: 10, P. 1 - 5

Published: Feb. 18, 2023

In our Daily life cycle, we come across many situations where being cautious and safe is the priority. Often these occur at a particular Defined place, like Hospitals, Airports, Food Sector, etc. Not only for Personal view but this can affect group of people too. The Fastest way to spread Bad Bacteria Viruses Airborne i.e. by contact with surrounding air. This includes significant aspect virus through CO2 released Human beings, Sneeze, coughing, At those times, wearing Mask comes in it best precaution. When preparation packaging, head caps or surgical are one identified asset. Wearing cap prevents hairs bacteria mix food. There Algorithms methodologies Provide Object Detection paper has Efficient working Approach. Our Model will detect beings Entrance Respective Areas Supervision Care should be taken without Fail. Screen show as well Voice Convey whether Person before entering Cautious place not.

Language: Английский

Citations

3
Automated Face Mask Detector Using Machine Learning: An Approach to Reduce Burden on Healthcare System DOI
Retinderdeep Singh, Chander Prabha

Published: Sept. 20, 2023

Taking the pandemic into consideration it is a prime step to work on prevention aspect. Although healthcare system breaking down due increased spread of COVID-19 its transmission by airborne route through cough and sneezing, urges need wear masks which includes personal protective equipment. Manual monitoring individuals at public area entry challenging part for administration so ease out this problem automated surveillance becomes hour. In current study, deep machine learning method used train model using an unstructured dataset various resources with sample size 1000 masked unmasked images individuals. The has undergo multiple layers phases like training phase, detection later providing E-commerce platform purchasing linking vending machine. results were achieved accuracy 99.8%, recall 99%, indicating that efficient in detecting face masks.

Language: Английский

Citations

3
Application Effect of Somatostatin Combined with Transnasal Ileus Catheterization in Patients with Acute Intestinal Obstruction and Advanced Gastric Cancer DOI Creative Commons
Zhenlu Li, Zhen Liu, Zongping Yu

et al.

Computational Intelligence and Neuroscience, Journal Year: 2022, Volume and Issue: 2022, P. 1 - 7

Published: June 11, 2022

Objective: To explore the application of somatostatin combined with nasal plug catheterization in patients advanced gastric cancer and acute intestinal obstruction. Methods. This study included 94 cases obstruction cancer, according to length hospital stay, were randomly divided into two groups: control group group, 47 each group. Based on observations made by team given treatment, we observed groups gastrointestinal function, serum index, quality life, therapeutic effect, adverse reactions. Results. Abdominal distention, abdominal pain duration, normal exhaust time significantly shorter than The was higher terms decompression volume, drainage circumference reduction within 24 hours (P < 0.05). After levels CRP, IgA, LPS, FABP lower before, former much those latter Compared before GIQLI scale score efficiency is lowers incidence postoperative complications Conclusion. For obstruction, it safe feasible use transnasal restore improve inflammatory response, promote improvement life high safety feasibility.

Language: Английский

Citations

4