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).
A square experimental arena with vegetation on one interior side was deployed in a Sharjah, United Arab Emirates desert. Individual darkling beetles (Coleoptera, Tenebrionidae) Akis subtricostata Redtenbacher, 1850 and Trachyderma philistina Reiche and Saulcy, 1857 were placed inside the arena at temperatures ranging between 27 - 49°C. Whether they chose the vegetated side of the arena or not was recorded, as well as how long it took for them to reach the vegetated side, if they chose it. Both species preferred the vegetated side at all temperatures, and the chance of them choosing the vegetated side increased significantly with increasing temperature (logistic regression, p = 0.0096 and p = 0.0003 for
... Show MoreThe OpenStreetMap (OSM) project aims to establish a free geospatial database for the entire world which is editable by international volunteers. The OSM database contains a wide range of different types of geographical data and characteristics, including highways, buildings, and land use regions. The varying scientific backgrounds of the volunteers can affect the quality of the spatial data that is produced and shared on the internet as an OSM dataset. This study aims to compare the completeness and attribute accuracy of the OSM road networks with the data supplied by a digitizing process for areas in the Baghdad and Thi-Qar governorates. The analyses are primarily based on calculating the portion of the commission (extra road) and
... Show MorePortland cement concrete is the most commonly used construction material in the world for decades. However, the searches in concrete technology are remaining growing to meet particular properties related to its strength, durability, and sustainability issue. Thus, several types of concrete have been developed to enhance concrete performance. Most of the modern concrete types have to contain supplementary cementitious materials (SCMs) as a partial replacement of cement. These materials are either by-products of waste such as fly ash, slag, rice husk ash, and silica fume or from a geological resource like natural pozzolans and metakaolin (MK). Ideally, the utilization of SCMs will enhance the concrete performance, minimize
... Show MoreThe purpose of this work is to study the classification and construction of (k,3)-arcs in the projective plane PG(2,7). We found that there are two (5,3)-arcs, four (6,3)-arcs, six (7,3)arcs, six (8,3)-arcs, seven (9,3)-arcs, six (10,3)-arcs and six (11,3)-arcs. All of these arcs are incomplete. The number of distinct (12,3)-arcs are six, two of them are complete. There are four distinct (13,3)-arcs, two of them are complete and one (14,3)-arc which is incomplete. There exists one complete (15,3)-arc.
Deep learning (DL) plays a significant role in several tasks, especially classification and prediction. Classification tasks can be efficiently achieved via convolutional neural networks (CNN) with a huge dataset, while recurrent neural networks (RNN) can perform prediction tasks due to their ability to remember time series data. In this paper, three models have been proposed to certify the evaluation track for classification and prediction tasks associated with four datasets (two for each task). These models are CNN and RNN, which include two models (Long Short Term Memory (LSTM)) and GRU (Gated Recurrent Unit). Each model is employed to work consequently over the two mentioned tasks to draw a road map of deep learning mod
... Show More