During COVID-19, wearing a mask was globally mandated in various workplaces, departments, and offices. New deep learning convolutional neural network (CNN) based classifications were proposed to increase the validation accuracy of face mask detection. This work introduces a face mask model that is able to recognize whether a person is wearing mask or not. The proposed model has two stages to detect and recognize the face mask; at the first stage, the Haar cascade detector is used to detect the face, while at the second stage, the proposed CNN model is used as a classification model that is built from scratch. The experiment was applied on masked faces (MAFA) dataset with images of 160x160 pixels size and RGB color. The model achieved lower computational complexity and number of layers, while being more reliable compared with other algorithms applied to recognize face masks. The findings reveal that the model's validation accuracy reaches 97.55% to 98.43% at different learning rates and different values of features vector in the dense layer, which represents a neural network layer that is connected deeply of the CNN proposed model training. Finally, the suggested model enhances recognition performance parameters such as precision, recall, and area under the curve (AUC).
One of the most important features of the Amazon Web Services (AWS) cloud is that the program can be run and accessed from any location. You can access and monitor the result of the program from any location, saving many images and allowing for faster computation. This work proposes a face detection classification model based on AWS cloud aiming to classify the faces into two classes: a non-permission class, and a permission class, by training the real data set collected from our cameras. The proposed Convolutional Neural Network (CNN) cloud-based system was used to share computational resources for Artificial Neural Networks (ANN) to reduce redundant computation. The test system uses Internet of Things (IoT) services through our cameras system to capture the images and upload them to the Amazon Simple Storage Service (AWS S3) cloud. Then two detectors were running, Haar cascade and multitask cascaded convolutional neural networks (MTCNN), at the Amazon Elastic Compute (AWS EC2) cloud, after that the output results of these two detectors are compared using accuracy and execution time. Then the classified non-permission images are uploaded to the AWS S3 cloud. The validation accuracy of the offline augmentation face detection classification model reached 98.81%, and the loss and mean square error were decreased to 0.0176 and 0.0064, respectively. The execution time of all AWS cloud systems for one image when using Haar cascade and MTCNN detectors reached three and seven seconds, respectively.
Image classification is the process of finding common features in images from various classes and applying them to categorize and label them. The main problem of the image classification process is the abundance of images, the high complexity of the data, and the shortage of labeled data, presenting the key obstacles in image classification. The cornerstone of image classification is evaluating the convolutional features retrieved from deep learning models and training them with machine learning classifiers. This study proposes a new approach of “hybrid learning” by combining deep learning with machine learning for image classification based on convolutional feature extraction using the VGG-16 deep learning model and seven classifiers. A hybrid supervised learning system that takes advantage of rich intermediate features extracted from deep learning compared to traditional feature extraction to boost classification accuracy and parameters is suggested. They provide the same set of characteristics to discover and verify which classifier yields the best classification with our new proposed approach of “hybrid learning.” To achieve this, the performance of classifiers was assessed depending on a genuine dataset that was taken by our camera system. The simulation results show that the support vector machine (SVM) has a mean square error of 0.011, a total accuracy ratio of 98.80%, and an F1 score of 0.99. Moreover, the results show that the LR classifier has a mean square error of 0.035 and a total ratio of 96.42%, and an F1 score of 0.96 comes in the second place. The ANN classifier has a mean square error of 0.047 and a total ratio of 95.23%, and an F1 score of 0.94 comes in the third place. Furthermore, RF, WKNN, DT, and NB with a mean square error and an F1 score advance to the next stage with accuracy ratios of 91.66%, 90.47%, 79.76%, and 75%, respectively. As a result, the main contribution is the enhancement of the classification performance parameters with images of varying brightness and clarity using the proposed hybrid learning approach.
This paper presents an improved technique on Ant Colony Optimization (ACO) algorithm. The procedure is applied on Single Machine with Infinite Bus (SMIB) system with power system stabilizer (PSS) at three different loading regimes. The simulations are made by using MATLAB software. The results show that by using Improved Ant Colony Optimization (IACO) the system will give better performance with less number of iterations as it compared with a previous modification on ACO. In addition, the probability of selecting the arc depends on the best ant performance and the evaporation rate.
Power stations must supply the electrical load demands to achieve optimal power system operation. To meet the future load, the power system dispatcher use load forecasting techniques to schedule unit generation resources. In this paper the short term load forecasting (STLF) using feed forward Artificial Neural Network (ANN) and Multiple Linear Regression (MLR) techniques for Iraqi power system (IPS) is presented. The ANN and MLR techniques are used to forecast one day ahead load for summer and winter season. The ANN gives a very small mean absolute percentage error (MAPE) compared with MLR but it takes a longer time for training process.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.