检索结果-内蒙古大学图书馆

Graph-Segmenter:graph transformer with boundary-aware attention for semantic segmentation

Frontiers of computer Science 2024年第5期18卷 97-108页

作者： Zizhang WU Yuanzhu GAN Tianhao XU Fan WANG Computer Vision Perception Department of ZongMu Technology Shanghai 201203China Faculty of Electrical Engineering Information TechnologyPhysicsTechnical University of BraunschweigBraunschweig 38106Germany

Thetransformer-based semantic segmentation approaches,which divide the image into different regions by sliding windows and model the relation inside each window,have achieved outstanding ***,since the relation modeling between windows was not the primary emphasis of previous work,it was not fully *** address this issue,we propose a Graph-Segmenter,including a graph transformer and a boundary-aware attention module,which is an effective network for simultaneously modeling the more profound relation between windows in a global view and various pixels inside each window as a local one,and for substantial low-cost boundary ***,we treat every window and pixel inside the window as nodes to construct graphs for both views and devise the graph *** introduced boundary-awareattentionmoduleoptimizes theedge information of the target objects by modeling the relationship between the pixel on the object's *** experiments on three widely used semantic segmentation datasets(Cityscapes,ADE-20k and PASCAL Context)demonstrate that our proposed network,a Graph Transformer with Boundary-aware Attention,can achieve state-of-the-art segmentation performance.

关键词： graph transformer graph relation network boundary-aware attention semantic segmentation

来源：评论

学校读者我要写书评

暂无评论

A Holistic Analysis and Experimental Testing of a Passive Heat Path for Electric Motors Slots

引用

IEEE Transactions on Transportation Electrification 2025年第3期11卷 8128-8138页

作者： Hebala, Ahmed Connor, Peter H. Nuzzo, Stefano Gerada, Chris Galea, Michael Arab Academy for Science Technology and Maritime Transport Electrical and Control Engineering department Alexandria5528341 Egypt Research Group Faculty of Engineering NottinghamNG7 2RD United Kingdom University of Modena and Reggio Emilia Department of Engineering Enzo Ferrari Modena41121 Italy University of Malta Department of Electrical Engineering MsidaMSD 2080 Malta

This article presents a thorough and broad analysis of the passive slot heat path (HP) concept. Six HP shapes are proposed and compared in terms of their combined thermal and electrical performance. Through an extensive search pool that covers a wide range of motor and slot designs and operating conditions, this article assesses the heat removal capabilities of the HP from the slot. The analysis is performed through a lumped parameter thermal network (LPTN) approach. Additionally, statistical analyses using correlations and linear regression models are carried out to improve understanding of this concept. Furthermore, a neural-network model is created to improve the regression model accuracy in predicting the slot temperature and temperature reduction. Finally, experimental testing and validation of the modeling are carried out on one of the six HP shapes. © 2015 IEEE.

关键词： Lumped parameter networks

来源：评论

学校读者我要写书评

暂无评论

Near-field Source Localization in 3-D Using Two Parallel Centrally Symmetric Unfold Coprime Array

引用

IEEE Transactions on Wireless Communications 2025年第6期24卷 4738-4749页

作者： Chen, Hua Li, Junjie Yang, Songjie Liu, Wei Eldar, Yonina C. Yuen, Chau Ningbo University Faculty of Electrical Engineering and Computer Science Ningbo315211 China University of Electronic Science and Technology of China National Key Laboratory of Wireless Communications Chengdu610054 China The Hong Kong Polytechnic University Department of Electrical and Electronic Engineering Kowloon Hong Kong Weizmann Institute of Science Faculty of Mathematics and Computer Science Rehovot7610001 Israel Nanyang Technological University School of Electrical and Electronics Engineering Singapore639798 Singapore

Most near-field (NF) localization algorithms cannot deal with the underdetermined case, while those which can are computationally expensive due to employment of fourth-order cumulants. In this work, a low-complexity solution is provided for underdetermined three-dimensional (3-D) NF localization, by employing second-order statistics with a tailored array configuration named two parallel centrally symmetric unfold coprime (TPSC) array. Its implementation can be divided into three stages. Firstly, the proposed algorithm constructs two cross-correlation matrices based on the received array data, which eliminates the non-linear range-related information of NF signals. Secondly, covariance and vectorization operations are applied to these two cross-correlation matrices to form a virtual array with extended aperture. Finally, the two-dimensional (2-D) angle parameters are estimated by the sparse and parametric approach (SPA) and a phase retrieval operation, and then the one-dimensional (1-D) range parameter is achieved by the multiple signal classification (MUSIC) algorithm. One specific feature is that the estimated angle and range parameters are matched automatically. An analysis of the properties of the TPSC array is provided, and an optimal parameter configuration is derived, given that the total number of array elements is fixed. Simulation results demonstrate that the designed TPSC array can achieve underdetermined 3-D NF localization, and deliver enhanced estimation capabilities, surpassing those of established algorithms. © 2002-2012 IEEE.

关键词： Multiple signal classification

来源：评论

学校读者我要写书评

暂无评论

Vision transformer with feature calibration and selective cross-attention for brain tumor classification

引用

Iran Journal of computer Science 2025年第2期8卷 335-347页

作者： Khaniki, Mohammad Ali Labbaf Mirzaeibonehkhater, Marzieh Manthouri, Mohammad Hasani, Elham Faculty of Electrical Engineering K. N. Toosi University of Technology Tehran Iran Department of Electrical and Computer Engineering Indiana University-Purdue University Indianapolis IN United States Faculty of Electrical and Electronic Engineering Shahed University Tehran Iran Department of Chemical and Environmental Engineering University of Cincinnati Cincinnati 45221 OH United States

Brain tumor classification is a challenging task in medical image analysis, with significant implications for patient diagnosis and treatment. The objective of this paper is to propose a novel approach to brain tumor classification using a Vision Transformer (ViT) with a novel cross-attention mechanism. Our approach leverages the strengths of transformers in modeling long-range dependencies and multi-scale feature fusion. We introduce two new mechanisms to improve the performance of the cross-attention fusion module: Feature Calibration Mechanism (FCM) and Selective Cross-Attention (SCA). The FCM calibrates the features from different branches to make them more compatible, while The SCA selectively attends to the most informative features. Our experimental results demonstrate that the proposed approach outperforms other state-of-the-art methods, including Convolutional Neural Networks (CNN), ViT, and Cross-attention ViT, achieving an accuracy of 98.93% and an F1-score of 98.83%. Furthermore, the addition of Stochastic Depth mechanism improves the accuracy to 99.24% and the F1-score to 99.23%. The proposed FCM and SCA mechanisms can be easily integrated into other ViT architectures, making them a promising direction for future research in medical image analysis. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024.

关键词： Attention mechanism Brain tumor classification Deep learning Medical imaging analysis Vision transformer

来源：评论

学校读者我要写书评

暂无评论

Enhanced Modified Contrast Enhancement and Exposure Fusion: a visionary dehazing approach for clarity in hazy imagery

引用

Multimedia Tools and Applications 2025年 1-24页

作者： Ayoub, Abeer El-Shafai, Walid El-Samie, Fathi E. Abd Hamad, Ehab K. I. EL-Rabaie, El-Sayed M. Department of Electronics and Electrical Communication Engineering Faculty of Electronic Engineering Menoufia University Menouf32952 Egypt Prince Sultan University Riyadh11586 Saudi Arabia Electrical Engineering Department Faculty of Engineering Aswan University Aswan81542 Egypt Department of Information Technology College of Computer and Information Sciences Princess Nourah Bint Abdulrahman University P.O. Box 84428 Riyadh11671 Saudi Arabia

In the field of image processing, haze is a type of atmospheric scattering that reduces contrast and clarity of images, frequently masking small details and far-off objects. It is mostly caused by microscopic airborne particles like smoke, dust, and water droplets that scatter incoming light. The scattered light, producing a hazy image, diminishes the direct light that reaches the camera sensor. Many image processing methods have been presened to solve this problem and lessen the impacts of haze. Usually, these methods work by determining how much haze is there in an image, and then using this information to get the image back to its assumed previous quality. In this work, we present the proposed Modified Contrast Enhancement and Exposure Fusion (MCEEF) technique. The MCEEF dehazing technique falls under the umbrella of enhancement-based dehazing techniques. In this technique, hazy frames or images undergo sharpening through a Smoothing-Sharpening Image Filter (SSIF) designed to accentuate the disparity between the haze and objects in images. Subsequently, Gamma Correction (GC) and Color Preserving Adaptive Histogram Equalization (CP-AHE) enhance the sharpened hazy images by augmenting their contrast. Dehazed images are then obtained by fusing the results of the CP-AHE- and GC-enhanced images. The MCEEF dehazing technique is aided with enhanced hazy images or frames as input for the dehazing process. Moreover, essential enhancement tools precede the MCEEF dehazing technique. These tools include homomorphic processing and Contrast Limited Adaptive Histogram Equalization (CLAHE), which are instrumental in controlling the dynamic range before the dehazing phase. In the proposed approach, a hazy image or frame is initially subjected to homomorphic processing, followed by the application of CLAHE, and finally, the MCEEF dehazing technique is employed. To demonstrate the effectiveness of the proposed approach, we apply it on both visible and Near Infrared (NIR) frames. We

关键词： Demulsification

来源：评论

学校读者我要写书评

暂无评论

Dual-Buck Bipolar Buck-Boost AC-AC Converter With Reduced Semiconductor Devices

引用

IEEE Transactions on Industrial Electronics 2024年第9期71卷 10495-10511页

作者： Ahmed, Hafiz Furqan Chien, Howard Khan, Ashraf Ali Islam, Shirazul Akbar, Fazal Alzaabi, Omar National Sun Yat-sen University Department of Electrical Engineering Kaohsiung80424 Taiwan Memorial University of Newfoundland Faculty of Engineering And Applied Science Department of Electrical And Computer Engineering St. John'sNLA1B 3X9 Canada Qatar University Department of Electrical Engineering Doha2713 Qatar University of Edinburgh EdinburghEH9 3JL United Kingdom Khalifa University Department of Electrical Engineering And Computer Science Advanced Power And Energy Center Abu Dhabi127788 United Arab Emirates

Dual-buck (DB) structured ac-ac converters are becoming advanced due to their inherent protection from open- and short-circuit risks, and elimination of commutation issue. However, the existing DB ac-ac converters provide restricted bipolar voltage operations with a large semiconductor device count and underutilized inductors. This article proposed a new reduced switch DB ac-ac converter without commutation issue. The propose converter provides multiple buck-boost voltage operations with noninverting and inverting outputs, making it a flexible topology for various ac voltage applications. The propose converter exhibits the continuous input and output currents irrespective of operation mode, and performs well with nonunity power factor loads. The positive and negative voltage gains of the proposed converter can address the grid voltage sag and swell issues, respectively, making it right candidate for application as dynamic voltage restorer. A detailed comparative evaluation is carried out with existing DB converters, which validated the effectiveness of the proposed converter, including;reduced semiconductor device count and their power ratings, reduced inductor sizes with improved magnetic utilization, enhanced voltage compensation ranges, and better efficiency. A laboratory scale hardware prototype of the proposed converter is built and thoroughly tested to validate its operation. © 1982-2012 IEEE.

关键词： Semiconductor devices

来源：评论

学校读者我要写书评

暂无评论

Analysis of the External Characteristics and Eddy Current Loss of PMSG-Based WECS Relying on a Multi-Domain Coupling Model 19th

Analysis of the External Characteristics and Eddy Current Lo...

引用

19th Annual Conference of China Electrotechnical Society, ACCES 2024

作者： Zuo, Peng Li, Weili Liu, Xiaoke Liu, Wenmao Li, Dan Qiao, Tianhuai School of Electrical Engineering Beijing Jiaotong University Beijing China Department of Electrical Engineering Tsinghua University Beijing China College of Computer and Control Engineering Northeast Forestry University Harbin China

ISBN: (纸本)9789819613861

With the trend towards larger-scale wind generators, the internal physical field and control strategy of high-capacity generator is becoming increasingly complex. The physical field change law inside the generator is often overlooked by the conventional wind energy conversion system (WECS) model. Therefore, developing a comprehensive and precise WECS model is essential for the safe operating of wind generators. Aiming at the above problems, this paper takes a 5 MW direct-drive permanent magnet synchronous generator (PMSG) as an example, and establishes a multi-domain coupling model. This approach can be used to investigate how the external characteristics of PMSG change under different operating conditions. Additionally, it has the ability to monitor how eddy currents are distributed in permanent magnets (PM), supporting generator design theoretically. © Beijing Paike Culture Commu. Co., Ltd. 2025.

关键词： Permanent magnets

来源：评论

学校读者我要写书评

暂无评论

Numerical Integration Study for Spherical Near-Field to Far-Field Transformations 12

Numerical Integration Study for Spherical Near-Field to Far-...

引用

12th International electrical engineering Congress, iEECON 2024

作者： Sukpreecha, Kitiphon Lertwiriyaprapa, Titipong Torrungrueng, Danai Akkaraekthalin, Prayoot Phaebua, Kittisak Faculty Of Technical Education Bangkok Thailand King Mongkut's University Of Technology North Bangkok Faculty Of Electrical Engineering Department Of Electrical And Computer Engineering Bangkok Thailand

ISBN: (纸本)9798350383591

This paper presents a numerical integration study for spherical near field (NF) to far field (FF) transformations. The trapezoidal and Simpson 1/3 numerical integration methods are employed in the NF to FF transformations algorithm. The performance of both numerical integrations is investigated. In this study, a seven patch array antenna is employed. Simulated FF results are performed by three dimensional (3D) simulation program (CST Microwave Studio) and used as the reference. The collected NF sampling data is divided into two cases;1) over-sampled NF angles of ? = f = 2° and 2) under-sampled NF angles of ? = f = 12°. The investigation reveals that, in the case of under-sampled near field (NF) angles of ? = f = 12°, the trapezoidal demonstrates lower errors of the FF radiation pattern in comparison to those of Simpson 1/3. Particularly, the trapezoidal is suitable for the under-sampled NF measurement case. © 2024 IEEE.

关键词： Microstrip antennas

来源：评论

学校读者我要写书评

暂无评论

Direct Drive Permanent Magnet Synchronous Generator: Design, Modeling, and control for Wind Energy Applications

Periodica polytechnica Electrical engineering and computer s...

引用

Periodica polytechnica electrical engineering and computer science 2024年第2期68卷 168-177页

作者： Larabi, Zina Ghedamsi, Kaci Aouzellag, Djamal Department of Electrical Engineering Faculty of Electrical Engineering and Computer Sciences University of Tizi-Ouzou P. O. B. 17 RP Tizi-Ouzou15000 Algeria Department of Electrical Systems Engineering Faculty of Technology University of Boumerdes Independence Avenue Boumerdes35000 Algeria Laboratoire de Maitrise des Energies Renouvelables Faculty of Technology University of Bejaia Road of Targa Ouzemmour Bejaia06000 Algeria

The prominent trend in wind turbine technology centers on the adoption of direct-drive permanent magnet synchronous generators (DD-PMSG), a choice driven by their capacity to deliver superior efficiency through the elimination of *** paper presents a comprehensive exploration of the design, modeling, and control aspects of a DD-PMSG intended for harnessing wind energy conversion. Initially, the geometrical design of the generator is meticulously carried out, adhering to predefined technical specifications and constraints. Subsequently, an in-depth internal modeling, focusing on the electromagnetic behavior of the designed generator, is executed using finite element analysis (FEA) through the Ansys Maxwell RMXpert software. Finally, the external modeling and control system integration of the designed generator, connected to the grid through power electronic converters, are simulated using the Matlab/Simulink software suite. The resulting findings underscore the efficacy and viability of the proposed generator for wind turbine applications, affirming its potential to enhance wind energy conversion systems. © 2024 Budapest University of Technology and Economics. All rights reserved.

关键词： Wind turbines

来源：评论

学校读者我要写书评

暂无评论

Digital Twin and TD3-Enabled Optimization of xEV Energy Management in Vehicle-to-Grid Networks

引用

IEEE Access 2025年 13卷 92495-92506页

作者： Saba, Irum Alobaidi, Abdulraheem H. Alghamdi, Sultan Tariq, Muhammad National University of Computer & Emerging Sciences Department of Electrical Engineering Islamabad44000 Pakistan King Abdulaziz University Faculty of Engineering Department of Electrical and Computer Engineering Jeddah21589 Saudi Arabia King Abdulaziz University Center of Research Excellence in Renewable Energy and Power Systems Jeddah21589 Saudi Arabia Princeton University Department of Electrical and Computer Engineering PrincetonNJ United States

The rapid expansion of extended electric vehicle (xEV) adoption necessitates optimizing energy storage systems (ESS) management for enhanced performance, longevity, and reliability. However, traditional ESS management approaches struggle with real-time state estimation, energy optimization, and predictive maintenance, leading to inefficiencies in battery utilization and sustainability. This paper addresses these challenges by proposing an advanced ESS framework that integrates digital twin (DT) technology with the twin-delayed deep deterministic policy gradient (TD3) algorithm, a state-of-the-art reinforcement learning method derived from the deep deterministic policy gradient (DDPG). This integration enables precise real-time estimation of critical ESS states, including state of charge (SoC), state of health (SoH), state of energy (SoE), and remaining useful life (RUL), thereby enhancing predictive maintenance and operational efficiency. The proposed framework facilitates proactive battery health monitoring, generates early warnings for potential failures, and enables intelligent battery swapping via DT-driven ESS control. By dynamically adapting ESS control strategies, the TD3 algorithm optimizes energy distribution, reduces energy consumption, and improves overall vehicle performance. With a prediction accuracy of 99.8% for SoC, SoH, and SoE, the proposed approach effectively addresses key challenges in xEV ESS management like cell balancing, dynamic charging rate adjustment, battery swapping decisions, and energy optimization, contributing to a more sustainable and efficient EV ecosystem. © 2013 IEEE.

关键词： Battery management systems

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：