Application of a Novel and Improved VGG-19 Network in the Detection of Workers Wearing Masks

Abstract

In order to work and travel safely during the outbreak of COVID-19, a method of security detection based on deep learning is proposed by using machine vision instead of manual monitoring. To detect the illegal behaviors of workers without masks in workplaces and densely populated areas, an improved convolutional neural network VGG-19 algorithm is proposed under the framework of tensorflow, and more than 3000 images are collected for model training and testing. Using VGG-19 network model, three FC layers are optimized into one flat layer and two FC layers with reduced parameters. The softmax classification layer of the original model is replaced by a 2-label softmax classifier. The experimental results show that the precision of the model is 97.62% and the recall is 96.31%. The precision of identifying the workers without masks is 96.82%, the recall is 94.07%, and the data set provided has a high precision. For the future social health and safety to provide favorable test data.

Explainable DCNN based chest X-ray image analysis and classification for COVID-19 pneumonia detection

“…Then, this deep CNN is used to distinguish bacterial from viral pneumonia amongst those patients with pneumonia at hospital with high predictive values using clinically relevant parts of the images. The CNN is designed based on VGG-19 37 which is a variant of VGG model which in short consists of 19 layers (16 convolution layers, 3 Fully connected layer, 5 MaxPool layers and 1 SoftMax layer as in Fig. 1 and reduced the levels, changing the convolutional kernels to make it more feasible.…”

“…The performance results of different CNN models (ResNet50V2 44 , InceptionV3 45 , VGG-16 46 , VGG-19 37 , DenseNet 47 ) are compared by using our COVID-19 dataset, as shown in Table 3 . For comparison testing, 300 chest X-ray images (100 images for each class) are used, learning rate is initialized as 0.001, epoch is set 120, and batch size is used as 8.…”

Hybrid Deep Learning Model for Endoscopic Lesion Detection and Classification Using Endoscopy Videos

“…In the recent past, many researchers used machine learning and deep learning to perform clinical tasks, such as analyzing a massive volume of unstructured data, processing images, identifying diseases, etc [ 3 , 12 , 19 , 23 , 27 ]. Medical imaging gained much attention from researchers, due to its importance in the world.…”

“…There are many segmentation and classification methods based on machine learning and computer-based techniques that are used in medical imaging. Whereas stomach disease detection is a favorite among all researchers, there are various stomach disease detection, segmentation, and classification methods that are already implemented, and every method consists of various steps, such as noise removal [ 28 ], segmentation [ 29 ], extraction of features [ 1 , 30 ], feature selection, feature fusion [ 31 ], and classification [ 1 , 15 , 27 ].…”

Untitled

“…Red neuronal convolucional compuesta por 16 capas convolucionales [43], tres fully-connected, cinco Max-Pool y una SoftMax, con un aproximado de 143 millones de parámetros. La arquitectura de VGG19 se aprecia en la Figura 11.…”