检索结果-内蒙古大学图书馆

Human Faces Detection and Tracking for Crowd Management in Hajj and Umrah

computers, Materials & Continua 2022年第6期71卷 6275-6291页

作者： Riad Alharbey Ameen Banjar Yahia Said Mohamed Atri Abdulrahman Alshdadi Mohamed Abid Department of Information Systems and Technology College of Computer Science and EngineeringUniversity of JeddahJeddahSaudi Arabia Electrical Engineering Department College of EngineeringNorthern Border UniversityArarSaudi Arabia Laboratory of Electronics and Microelectronics(LR99ES30) Faculty of Sciences of MonastirUniversity of MonastirTunisia College of Computer Sciences King Khalid UniversityAbhaSaudi Arabia CES Laboratory ENISUniversity of SfaxTunisia

Hajj and Umrah are two main religious duties for *** help faithfuls to perform their religious duties comfortably in overcrowded areas,a crowd management system is a must to control the entering and exiting for each *** the number of people is very high,an intelligent crowd management system can be developed to reduce human effort and accelerate the management *** this work,we propose a crowd management process based on detecting,tracking,and counting human faces using Artificial Intelligence *** detection and counting will be performed to calculate the number of existing visitors and face detection and tracking will be used to identify all the humans for security *** proposed crowd management system is composed form three main parts which are:(1)detecting human faces,(2)assigning each detected face with a numerical identifier,(3)storing the identity of each face in a database for further identification and *** main contribution of this work focuses on the detection and tracking model which is based on an improved object detection *** improved Yolo v4 was used for face detection and *** has been very effective in detecting small objects in highresolution *** novelty contained in thismethod was the integration of the adaptive attention mechanism to improve the performance of the model for the desired *** wise attention mechanism was applied to the output layers while both channel wise and spatial attention was integrated in the building *** main idea from the adaptive attention mechanisms is to make themodel focus more on the target and ignore false positive *** demonstrated the efficiency of the proposed method through expensive experimentation on a publicly available *** wider faces dataset was used for the train and the evaluation of the proposed detection and tracking *** proposed model has achieved good results with 91.2%of mAP and a processing speed

关键词： Crowdmanagement Hajj and Umrah face detection object tracking convolutional neural networks(CNN) adaptive attention mechanisms

来源：评论

学校读者我要写书评

暂无评论

Exploring rhythmic visual cue impacts on gait initiation using functional EEG source connectivity 8

Exploring rhythmic visual cue impacts on gait initiation usi...

引用

8th International Conference on Intelligent Computing and Signal Processing, ICSP 2023

作者： Yan, Wenfeng Zhou, Huilin Wu, Jiajia Jiang, Wensong Zuo, Guokun Shi, Changcheng Faculty of Electrical Engineering and Computer Science Ningbo University Zhejiang Ningbo315211 China Chinese Academy of Sciences Ningbo Institute of Materials Technology and Engineering Zhejiang Ningbo315201 China Ningbo Cixi Institute of Biomedical Engineering Zhejiang Ningbo315300 China Chinese Academy of Sciences Zhejiang Engineering Research Center for Biomedical Materials Ningbo Institute of Materials Technology and Engineering Zhejiang Ningbo315300 China Wenzhou Medical University Cixi Biomedical Research Institute Zhejiang China

ISBN: (纸本)9798350302455

The underlying neural mechanisms of rhythmic visual cues to improve gait freezing symptoms are still remain unclear. Previous studies of rhythmic visual cues have not taken into account the fact that the brain processes information as a result of the interplay between various brain regions. This study applied the EEG source connectivity method to investigate the cortical response and to evaluate the functional integration relationship of the brain networks under the non-rhythmic and rhythmic visual cues. Results showed that there were more network connectivity under the rhythmic visual cues, which suggested that rhythmic visual cues strengthen the brain's visual processing and cognitive control functions by enhancing the interactions between brain regions. In addition, the brain functional network under rhythmic visual cues possessed stronger functional separation and functional integration. This study suggested that the modulatory mechanism of rhythmic visual cues on gait initiation may be related to the enhanced information processing efficiency of the brain functional network through visual information input. The innovative source localization of EEG signals and the combination of the analysis of functional brain networks complement the methodology of the study of rhythmic visual cues, thus filling a gap in this research. © 2023 IEEE.

关键词： Gait analysis

来源：评论

学校读者我要写书评

暂无评论

Fine-tuning Specialized NER Model for Symptom Extraction from Slovenian Medical Texts

Fine-tuning Specialized NER Model for Symptom Extraction fro...

引用

International Conference on Mathematics and computers in sciences and in Industry (MCSI)

作者： Rigon Sallauka Umut Arioz Izidor Mlakar Laboratory of Digital Signal Processing University of Maribor Faculty of Electrical Engineering and Computer Science Maribor Slovenia

ISBN: (数字)9798350377224

ISBN: (纸本)9798350377231

In this study, we aim to develop a robust model capable of accurately extracting symptoms from clinical texts written in the Slovenian language. Our methodology involves several steps to achieve this objective. Initially, we identify datasets containing clinical texts in English and select a specialized model designed for symptom extraction from these texts. Then, we annotate the datasets and undertake the translation process into Slovenian, ensuring the preservation of annotations to maintain data integrity. Utilizing the translated dataset, we proceed to fine-tune the selected model specifically for Slovenian texts. Despite our efforts, optimal performance has not yet been attained. Challenges encountered include the translation methodology employed and the architectural limitations of the chosen model, which is primarily optimized for English texts. However, ongoing efforts are underway to address these challenges and enhance the model's performance. Our study represents a step towards improving automated symptom extraction in Slovenian clinical texts, with ongoing work focused on mitigating identified issues to achieve better outcomes.

关键词： Adaptation models Translation Accuracy Text analysis Annotations Computational modeling Data models Mathematical models Reliability Tuning

来源：评论

学校读者我要写书评

暂无评论

Weak and Extremely Low Frequency Magnetic Fields Alter the Growth Rates of HT-1080 Fibrosarcoma

Weak and Extremely Low Frequency Magnetic Fields Alter the G...

引用

US National Committee of URSI National Radio science Meeting (USNC-URSI NRSM)

作者： Nhat Dang Jason Keller Marek Bajtos Hakki Gurhan Frank Barnes Department of Electrical Computer and Energy Engineering College of Engineering and Applied Science University of Colorado Boulder Boulder Colorado USA Department of Electromagnetics and Biomedical Engineering Faculty of Electrical Engineering and Information Technology University of Zilina Zilina Slovakia

ISBN: (数字)9781946815200

ISBN: (纸本)9798331513979

This study presents experimental results showing that weak, extremely low frequency electromagnetic (EM) fields influence the growth rates of HT-1080 fibrosarcoma cells in vitro and offers a theoretical explanation for these effects. The results demonstrate that weak magnetic fields both increase and decrease the growth rates of fibrosarcoma cells by altering the mitochondrial superoxide $(\mathrm{O}^{2-})$ and calcium ions $(\text{Ca}^{2+})$ concentrations in living cells. Over a 4-day exposure period, fibrosarcoma growth rates were modulated in the presence of low-level alternating current (AC) magnetic fields at a $10 \mu \mathrm{T}$ amplitude at 13.5 Hz to 30 Hz frequency, superimposed on a $45 \mu \mathrm{T}$ static background field. Effects due to changes in amplitude from 10 to $30 \mu \mathrm{T}$ at selected frequencies are also investigated with the same superimposed static field.

关键词： Frequency modulation Calcium Ions Magnetic fields In vitro Electromagnetics

来源：评论

学校读者我要写书评

暂无评论

Real-Time Sensor Fault Tolerant Control of DC-DC Converters in DC Microgrids Using a Switching Unknown Input Observer

引用

IEEE Access 2025年 13卷 95837-95850页

作者： Ouahabi, Mohammed Said Benyounes, Abdelhafid Barkat, Said Ihammouchen, Syphax Rekioua, Toufik Rabehi, Abdelaziz El-Kenawy, El-Sayed M. Alharbi, Amal H. University of M’sila Electrical Engineering Laboratory Faculty of Technology M’sila28000 Algeria University of Béjaïa Laboratoire de Technologie Industrielle et de l’Information Béjaïa06000 Algeria Faculty of Science and Technology Djelfa17000 Algeria Isa Town11937 Bahrain Applied Science Private University Applied Science Research Center Amman11931 Jordan Princess Nourah bint Abdulrahman University College of Computer and Information Sciences Department of Computer Sciences P.O. Box 84428 Riyadh11671 Saudi Arabia

DC Microgrids are rapidly developing, driven by the increasing integration of renewable sources and dynamic loads. Their growing adoption in critical applications raises the need for operational safety and reliability, particularly under fault conditions. This paper presents a sensor fault-tolerant control (FTC) for DC-DC power converters in DC Microgrids designed to maintain resilient and reliable operation. The proposed FTC leverages a single Switching Unknown Input Observer (SW-UIO) at the bottom control level to estimate arbitrary sensor faults with a high accuracy (0.3% fault estimation error), allowing a real-time identification and precise control reconfiguration to accommodate sensor faults with minimal computational burden. The approach is validated under diverse conditions, including time-varying loads, complete sensor failure, noise abnormality, and simultaneous faults/failure, achieving a consistent fault accommodation of 1 μs from fault occurrence to compensation. Real-time simulation has been performed in the OPAL-RT environment to assess the efficacy and superiority of the proposed sensor fault tolerant strategy in comparison to reference FTC strategies. © 2013 IEEE.

关键词： Identification (control systems)

来源：评论

学校读者我要写书评

暂无评论

Low-Cost, High-Performance Algorithm and Hardware Architecture for SAO Mode Decision 2

Low-Cost, High-Performance Algorithm and Hardware Architectu...

引用

2nd IEEE International Conference on Signal, Information and Data Processing, ICSIDP 2024

作者： Hou, Bingjing Liu, Chang Zhang, Yuejun Huang, Leilei Jing, Minge Fan, Yibo Fudan University State Key Laboratory of Integrated Chips and Systems Shanghai China Ningbo University Faculty of Electrical Engineering and Computer Science Ningbo China East China Normal University Institute of Microelectronic Circuits and Systems Shanghai China

ISBN: (纸本)9798331515669

Sample adaptive offset (SAO) is applied for reducing sample distortion and attenuating ringing artifacts in both HEVC and VVC standards. The rate-distortion optimization process is used to select the best SAO parameter. It accounts for the majority of SAO complexity and determines the SAO filtering performance. Therefore, the rate-distortion cost calculation should be well handled to achieve both low cost and high coding performance. In this paper, we propose a simplified and optimized rate estimation algorithm based on the characteristics of SAO parameter syntax and entropy coding process. And we propose a hierarchical flow for rate-distortion cost calculation to further reduce the computational cost and storage cost for SAO mode decision hardware. The results show that the proposed algorithm achieves an average YUV loss of -0.10%, and the proposed SAO mode decision hardware architecture can be implemented by only 21.8k gates using TSMC 65nm process under 400 MHz. © 2024 IEEE.

关键词： Electric distortion

来源：评论

学校读者我要写书评

暂无评论

Contrast enhancement of optical videoscope images

引用

Iran Journal of computer science 2023年第3期6卷 261-275页

作者： Ammar, Reda Aboshosha, Sahar El-Hag, Noha A. Saifeldeen, Reda El-Shafai, Walid El-Banby, Ghada M. Ashiba, Huda I. Khalaf, Ashraf A. M. Elazm, Atef Abou El-Rabaie, El-Sayed M. Abd El-Samie, Fathi E. El-Safrawey, Amir Department of Electronics and Electrical Communications Engineering Faculty of Electronic Engineering Menoufia University Menouf 32952 Egypt Department of Electronics and Electrical Comm. Faculty of Engineering Minia University Minya Egypt Department of Management of Information Systems Higher Institute of Commercial Science El-Mahala El-Kobra Egypt Department of Industrial Electronics and Control Engineering Faculty of Electronic Engineering Menoufia University Menouf 32952 Egypt Department of Electronics and Electrical Communications Engineering Bilbis Higher Institute of Engineering Sharqia Bilbis Egypt Security Engineering Laboratory Computer Science Department Prince Sultan University Riyadh 11586 Saudi Arabia Department of Information Technology College of Computer and Information Sciences Princess Nourah Bint Abdulrahman University P.O. Box 84428 Riyadh 11671 Saudi Arabia

The videoscope (VS) images have poor quality and low contrast. Hence, in this paper, three proposed frameworks to improve the quality of VS images are presented. The first framework depends on contrast-limited adaptive histogram equalization (CLAHE) with adaptive gamma correction. The second framework depends on homomorphic processing. The third framework is based on the homomorphic enhancement with histogram equalization. The first one is superior from the entropy perspective, while the third one is superior from the contrast perspective. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2023.

关键词： Fuzzy enhancement Histogram equalization Homomorphic enhancement Spectral entropy

来源：评论

学校读者我要写书评

暂无评论

An Enhanced Control Strategy for Standalone Brushless Doubly Fed Induction Generator Under Unbalanced Load Using Reactive Power Loop 1

An Enhanced Control Strategy for Standalone Brushless Doubly...

引用

1st IEEE IAS Global Conference on Emerging Technologies, GlobConET 2022

作者： Elbabo Mohammed, Omer Mohammed Liu, Yi Xu, Wei Islam, Md. Rabiul University of Sinnar Faculty of Engineering Department of Electrical Engineering Sinnar Sudan Huazhong University of Science and Technology State Key Laboratory of Advanced Electromagnetic Engineering and Technology School of Electrical and Electronic Engineering Wuhan China University of Wollongong School of Electrical Computer and Telecommunications Engineering Wollongong Australia

ISBN: (数字)9781665443579

ISBN: (纸本)9781665443579

Without brushes and slip rings, the brushless double-fed induction generator (BDFIG) can operate more stably and reliably, so it is prospective to be one of important machines for standalone systems in the next few years. The stability of amplitude of voltage and frequency of power winding (PW) is major aim of control in off-grid system. Also, the standalone system is oversensitive to abnormal running situations, especially with unbalanced conditions. The unbalanced loads would lead to harmful phenomena, such as the unbalanced impact of current and voltage of PW. To handle this issue, the reactive power loop is presented to minimize voltage unbalance of PW through creating control winding (CW) compensation voltage via converter of machine side (MSC). The simulation is executed to test the performance of presented method. © 2022 IEEE.

关键词： Reactive power

来源：评论

学校读者我要写书评

暂无评论

Deep Phase Mapping 3D based on Single-Frame Digital Moiré Fringe 4

Deep Phase Mapping 3D based on Single-Frame Digital Moiré F...

引用

4th International Conference on Electronic Information engineering and computer Communication, EIECC 2024

作者： Di, Rui Hu, Hailing Hu, Meili Song, Baoan Ningbo University Faculty of Electrical Engineering and Computer Science Ningbo315211 China Ningbo University Key Laboratory of Photoelectric Detection Materials and Devices of Zhejiang Province Ningbo315211 China Engineering Research Center for Advanced Infrared Photoelectric Materials and Devices of Zhejiang Province Ningbo315211 China

ISBN: (纸本)9798331534622

To enhance three-dimensional(3D) measurement precision with minimal fringes, the deep phase mapping 3D based on single-frame digital moiré fringe method (DDM) is proposed. By projecting just one high-frequency fringe and combining wrapped phase with fringe order to calculate continuous phase, it achieves 3D reconstruction through training on three process variables, thereby enhancing measurement precision. Experimental DDM measurements of 3D contours for fox and complex face models surpassed traditional single-frame methods in accuracy. Further evaluation on standard planes confirmed DDM's precision through lower RMSE (root mean square error) and MAE (mean absolute error) values, underlining its superior reconstruction accuracy. It holds great promise for its potential applications in real-time industrial measurements. © 2024 IEEE.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

Incident Detection with Pruned Residual Multilayer Perceptron Networks

Incident Detection with Pruned Residual Multilayer Perceptro...

引用

Federated Conference on computer science and Information Systems (FedCSIS)

作者： Mohamad Soubra Marek-Kisiel Dorohinicki Marcin Kurdziel Marek Zachara Faculty of Computer Science Electronics and Telecommunications AGH University Krakow Poland Faculty of Electrical Engineering Automatics Computer Science and Biomedical Engineering AGH University Krakow Poland

Internet of things (IoT) has opened new horizons in connecting all sorts of devices to the internet. However, continuous demand for connectivity increases the cybersecurity risks, rendering IoT devices more prone to cyberattacks. At the same time, rapid advances in Deep Learning (DL)-based algorithms provide state-of-the-art results in many classification tasks, including classification of network traffic or system logs. That said, deep learning algorithms are considered computationally expensive as they require substantial processing and storage capacity. Sadly, IoT devices have limited resources, making renowned DL models hard to implement in this environment. In this paper we present a Residual Neural Network inspired DL-based Intrusion Detection System (IDS) that incorporates weight pruning to make the model more compact in size and resource consumption. Additionally, the proposed system leverages feature selection algorithms to reduce the feature-space size. The model was trained on the NSL-KDD dataset benchmark. Experimental results show that the proposed system is effective, being able to classify network traffic with an F 1 score of up to 98.9% before the pruning and an F 1 score of up to 97.5% after pruning 90% of network weights.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：