检索结果-内蒙古大学图书馆

A Hybrid Deep Learning-ViT Model and A Meta-Heuristic Feature Selection Algorithm for Efficient Remote Sensing Image Classification

引用

International Journal of Computational Intelligence systems 2025年第1期18卷 1-37页

作者： Ahmed, Bilal Naqvi, Syed Rameez Akram, Tallha Peng, Lu Almarshad, Fahdah Department of Electrical and Computer Engineering COMSATS University Islamabad Wah Campus Wah Cantonment 47040 Pakistan Department of Computer Science Tulane University New Orleans 70118 LA United States Department of Information Systems College of Computer Engineering and Sciences Prince Sattam bin Abdulaziz University Al-Kharj 11942 Saudi Arabia

Recent deep learning techniques driven by large datasets demonstrate the significant impact of feature learning in remote sensing for land use and cover classification, particularly exemplified by CNNs. While the pre-trained models showed good classification performance, they struggled to classify remote-sensing images with high precision accurately. In this study, we introduced XNANet, a self-attention-based CNN network for image classification. Bayesian optimization has been used to initialize the hyperparameters of the proposed model to improve training on the radiographic images. We suggested a novel network-level approach via the fusion of deep structure utilizing tiny-32 ViT and XNANet. For the first time, the tiny-32 vision transformer architecture has been utilized for RS images and combined with XNANet through network-level fusion. Following the fusion process, the model focused on RS image datasets and obtained deep features from the self-attention layer. The features that have been extracted are subject to selection, utilizing a novel meta-heuristic feature selection algorithm, RF-DE. The selected features are categorized using three popular classifiers. The proposed architecture’s experimental process was executed on the AID, RSSCN7, and SIRI-WHU datasets, resulting in accuracies of 98.9%, 99.3%, and 99.7%, respectively. Similarly, RF-DE was evaluated against six popular feature selection algorithms, yielding accuracies of 98.9%, 99.3%, and 99.7%, respectively. An in-depth statistical analysis was conducted to evaluate the suggested ensemble and RF-DE and demonstrate that the fusion model attained enhanced accuracy with RF-DE. Furthermore, recent techniques and proposed methods are compared, demonstrating enhanced precision, recall, and accuracy. © The Author(s) 2025.

关键词： Attention mechanism Bio-inspired feature selection CNN architecture Image classification Red Fox algorithm Remote sensing Vision transformer

来源：评论

学校读者我要写书评

暂无评论

Study of orography–MJO interaction and its impact on rainfall variability in West Sumatra

引用

Journal of Earth System Science 2025年第2期134卷 1-27页

作者： Satiadi, Didi Purwaningsih, Anis Harjupa, Wendi Saufina, Elfira Fathrio, Ibnu Trismidianto Rahmatia, Fahmi Pratama, Ridho Hashiguchi, Hiroyuki Shimomai, Toyoshi Research Center for Climate and Atmosphere National Research and Innovation Agency (BRIN) Bandung Indonesia Department of Computer Engineering School of Electrical Engineering Telkom University Bandung Indonesia Disaster Prevention Research Institute Kyoto University Kyoto Japan Research Institute for Sustainable Humanosphere Kyoto University Kyoto Japan Department of Electronic and Control Systems Engineering Shimane University Shimane Japan

Abstract: This study examines the impact of orography and the Madden–Julian Oscillation (MJO) on rainfall patterns in West Sumatra, Indonesia, using ERA5 reanalysis data. We analyzed variables such as precipitation, zonal and vertical winds, vertically integrated moisture convergence (VIMC), Froude number, lifting condensation level (LCL), level of free convection (LFC), level of non-buoyancy (LNB), convective available potential energy (CAPE), and convective inhibition (CINH), focusing on MJO phases 3 (MJO-3) and 4 (MJO-4). The ERA5 data was validated with observations from the Agam Observation Station, including data from the equatorial atmosphere radar, X-band Doppler Radar, optical rain gauge, microwave radiometer, and radiosonde. Our analysis, covering 22 years (2001–2022) of hourly data from five locations at varying elevations, reveals that topography significantly influences rainfall distribution, with windward slopes and mountain peaks receiving the most rainfall. This pattern closely aligns with VIMC, indicating that moisture convergence is a key driver of rainfall variability. MJO-3 significantly enhances rainfall by reducing zonal wind speeds, thereby increasing moisture convergence, while MJO-4 has a minimal impact, with increased zonal wind speeds limiting moisture convergence. The study concludes that VIMC patterns, influenced by wind dynamics, are central to understanding rainfall variability in West Sumatra’s mountainous terrain under different MJO phases. Research highlights: Orography strongly influences rainfall in West Sumatra, with the most precipitation on windward slopes and peaks, linked to vertically integrated moisture convergence (VIMC). MJO-3 boosts rainfall by reducing zonal wind speeds, which enhances moisture convergence and emphasizes wind dynamics in rainfall variability. MJO-4 has minimal impact on rainfall, as increased zonal winds restrict moisture convergence, leading to slight changes in precipitation. VIMC is crucial for under

关键词： MJO Orography rainfall topography West Sumatra

来源：评论

学校读者我要写书评

暂无评论

Normative Safety Regulations for Collaborative Robots

Normative Safety Regulations for Collaborative Robots

引用

IEEE/SICE International Symposium on System Integration

作者： Laura Geiger J. Rogelio Guadarrama-Olvera Gordon Cheng Department of Electrical and Computer Engineering Chair for Cognitive Systems Technische Universität München Munich Germany

ISBN: (数字)9798331531614

ISBN: (纸本)9798331531621

Collaborative robots (cobots) are designed to operate alongside humans in shared workspaces, enhancing efficiency and flexibility. Like all machines and robots, they are regulated by the EU Machinery Regulation and must bear the CE marking before market introduction. While the regulation provides broad and general safety requirements, harmonized standards such as ISO 12100, ISO 13849, and IEC 62061 offer more specific guidance for achieving safety in machinery design. For industrial robots, ISO 10218 outlines safety requirements, with the technical specification ISO/TS 15066 further addressing unique considerations of collaborative industrial robots. By providing a comprehensive overview of these regulations and standards, this paper aims to guide manufacturers through the critical steps required for achieving legal compliance and ensuring safe market introduction of cobots in Europe.

关键词： Service robots Law ISO Standards Collaborative robots System integration Industrial robots Regulation Safety Machinery Standards

来源：评论

学校读者我要写书评

暂无评论

Effect of Magnetic Couplings on Conducted EMI of GaN-Based PFC Converter

Effect of Magnetic Couplings on Conducted EMI of GaN-Based P...

引用

Annual IEEE Conference on Applied Power Electronics Conference and Exposition (APEC)

作者： Tyler McGrew Qiang Li The Bradley Department of Electrical and Computer Engineering Center for Power Electronic Systems Virginia Tech Blacksburg VA USA

ISBN: (数字)9798331516116

ISBN: (纸本)9798331516123

Wide-bandgap power devices have enabled power supplies to increase in density while maintaining high efficiency. However, EMI issues remain one of the most challenging aspects when designing a power converter. These issues are often caused by parasitic near-field couplings which can degrade or bypass the attenuation of the EMI filter. This paper investigates how parasitic magnetic couplings impact the conducted noise of a high-frequency PFC converter using PCB-winding inductor. Analytical models are presented for both differential mode (DM) and common mode (CM) noise, including the effects of magnetic couplings between the PFC inductor and EMI filter. Then, three ferrite inductor core designs are evaluated to demonstrate how the core’s geometry and permeability affect its leakage magnetic field. Choosing the optimal core design is shown to significantly reduce the magnetic field outside the PFC inductor, which reduces DM noise by more than 20 dB and reduces CM noise by 7 dB.

关键词： Couplings Ferrites Magnetic cores Magnetic separation Atmospheric modeling Electromagnetic interference Noise Air gaps Permeability Inductors

来源：评论

学校读者我要写书评

暂无评论

Small-Signal Analysis and External Ramp Design for Multiphase Current-Mode Constant On-Time Control with Phase Overlapping

Small-Signal Analysis and External Ramp Design for Multiphas...

引用

Annual IEEE Conference on Applied Power Electronics Conference and Exposition (APEC)

作者： Sundaramoorthy Sridhar Qiang Li The Bradley Department of Electrical and Computer Engineering Center for Power Electronic Systems Virginia Tech Blacksburg VA USA

ISBN: (数字)9798331516116

ISBN: (纸本)9798331516123

High-current processor core voltage regulators (VRs) use current-mode constant on-time (COT) control to simplify phase interleaving and achieve high control bandwidth. Reducing VR bus voltage and increasing phase count results in steady-state phase overlapping within their operating duty range. With phase overlap, a smaller external ramp reduces the inner current loop stability margin and introduces large magnitude peaks in control-to-output response. These peaks could lead to double crossover, reduce gain margin, and make high-bandwidth outer voltage loop unstable in phase overlapping regions. Increasing the ramp damps these peaks and improves the gain margin but reduces the phase margin at the target bandwidth. Hence, this paper simplifies infinite-order describing function model and provides external ramp design guidelines to attain desired stability margin at the target bandwidth. The simplified model and small-signal analysis are validated using SIMPLIS simulation and experiments from a six-phase current-mode COT buck converter.

关键词： Regulators Design methodology Transfer functions Process control Bandwidth Stability analysis Resonance Power electronics Steady-state Voltage control

来源：评论

学校读者我要写书评

暂无评论

Soft Sensory Socks Measure Contact Forces During Locomotion

Soft Sensory Socks Measure Contact Forces During Locomotion

引用

IEEE/SICE International Symposium on System Integration

作者： John Nassour Jan-Erik Hühne Jin-Ho Lee Gordon Cheng Department of Electrical and Computer Engineering Chair for Cognitive Systems Technische Universität München Munich Germany

ISBN: (数字)9798331531614

ISBN: (纸本)9798331531621

Gait monitoring systems play a crucial role in guiding rehabilitation, improving athletic performance, preventing injuries, and ensuring overall mobility and quality of life. Most of these systems are restricted to specialized setups and require calibrations. We propose a soft sensor design using silicone tubes mounted on the bottom side of socks to measure ground reaction force (GRF) during locomotion. Three sensors were mounted on the socks’ anterior, lateral, and posterior areas on the left and right feet. The sensory socks enable the analysis of gait while walking and running at different speeds while wearing shoes and also barefoot. The proposed soft sensory socks open the door to analyzing locomotion in different environments and also barefoot. Preliminary gait analyses confirm the different gait parameters observed in the biomechanics of barefoot locomotion.

关键词： Legged locomotion Force measurement Soft sensors Force Footwear System integration Gait recognition Monitoring Injuries Foot

来源：评论

学校读者我要写书评

暂无评论

Material-Driven Mechanical Programming of Soft Robotic Tentacles

Material-Driven Mechanical Programming of Soft Robotic Tenta...

引用

IEEE/SICE International Symposium on System Integration

作者： Yao Lu Asmaa Amraouza Yifei Peng Dan Li John Nassour Gordon Cheng Department of Electrical and Computer Engineering Chair for Cognitive Systems Technische Universität München Munich Germany

ISBN: (数字)9798331531614

ISBN: (纸本)9798331531621

Soft robotic grippers with stochastic and topological grasping capabilities can be highly desirable for gentle contact and interaction with fragile objects of various shapes. In this study, we developed soft tentacles using silicon elastomeric material with an embedded pneumatic channel, which generates desirable three-dimensional (3D) curling deformation under pneumatic pressure. Additionally, various ratios and geometrical distributions of two silicon materials, which differ primarily in stiffness, were investigated to determine their effect on grasping efficiency. The tentacle's grasping performance was systematically tested across multiple tentacle designs and grasping strategies. Results showed that optimizing the combination of softer elastomers with stiffer materials significantly improved the tentacle's ability to securely grip and carry loads while maintaining gentle contact with objects (up to 5 kg). The pneumatic tentacle's simple control mechanism and versatility in handling objects of various shapes and sizes offer a low-cost, adaptable solution for future applications in soft robotics.

关键词： Three-dimensional displays Shape Deformation Grasping Elastomers System integration Soft robotics Skin Silicon Filling

来源：评论

学校读者我要写书评

暂无评论

MIRAM: Masked Image Reconstruction Across Multiple Scales for Breast Lesion Risk Prediction

arXiv

引用

arXiv 2025年

作者： Vo, Hung Q. Yuan, Pengyu Yin, Zheng Wong, Kelvin K. Ezeana, Chika F. Ly, Son T. Wong, Stephen T.C. Nguyen, Hien V. Department of Electrical and Computer Engineering University of Houston United States Systems Medicine and Biomedical Engineering Houston Methodist United States

Self-supervised learning (SSL) has garnered substantial interest within the machine learning and computer vision communities. Two prominent approaches in SSL include contrastive-based learning and self-distillation utilizing cropping augmentation. Lately, masked image modeling (MIM) has emerged as a more potent SSL technique, employing image inpainting as a pretext task. MIM creates a strong inductive bias toward meaningful spatial and semantic understanding. This has opened up new opportunities for SSL to contribute not only to classification tasks but also to more complex applications like object detection and image segmentation. Building upon this progress, our research paper introduces a scalable and practical SSL approach centered around more challenging pretext tasks that facilitate the acquisition of robust features. Specifically, we leverage multi-scale image reconstruction from randomly masked input images as the foundation for feature learning. Our hypothesis posits that reconstructing high-resolution images enables the model to attend to finer spatial details, particularly beneficial for discerning subtle intricacies within medical images. The proposed SSL features help improve classification performance on the Curated Breast Imaging Subset of Digital Database for Screening Mammography (CBIS-DDSM) dataset. In pathology classification, our method demonstrates a 3% increase in average precision (AP) and a 1% increase in the area under the receiver operating characteristic curve (AUC) when compared to state-of-the-art (SOTA) algorithms. Moreover, in mass margins classification, our approach achieves a 4% increase in AP and a 2% increase in AUC. © 2025, CC BY-NC-SA.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Accelerating OTA Circuit Design: Transistor Sizing Based on a Transformer Model and Precomputed Lookup Tables

Accelerating OTA Circuit Design: Transistor Sizing Based on ...

引用

Design, Automation and Test in Europe Conference and Exhibition

作者： Subhadip Ghosh Endalk Y. Gebru Chandramouli V. Kashyap Ramesh Harjani Sachin S. Sapatnekar Department of Electrical and Computer Engineering University of Minnesota Minneapolis MN USA Cadence Design Systems Portland OR USA

ISBN: (数字)9783982674100

ISBN: (纸本)9798331534646

Device sizing is crucial for meeting performance specifications in operational transconductance amplifiers (OTAs), and this work proposes an automated sizing framework based on a transformer model. The approach first leverages the driving-point signal flow graph (DP-SFG) to map an OTA circuit and its specifications into transformer-friendly sequential data. A specialized tokenization approach is applied to the sequential data to expedite the training of the transformer on a diverse range of OTA topologies, under multiple specifications. Under specific performance constraints, the trained transformer model is used to accurately predict DP-SFG parameters in the inference phase. The predicted DP-SFG parameters are then translated to transistor sizes using a precomputed look-up table-based approach inspired by the 9m / I d methodology. In contrast to previous conventional or machine-learning-based methods, the proposed framework achieves significant improvements in both speed and computational efficiency by reducing the need for expensive SPICE simulations within the optimization loop; instead, almost all SPICE simulations are confined to the one-time training phase. The method is validated on a variety of unseen specifications, and the sizing solution demonstrates over 90% success in meeting specifications with just one SPICE simulation for validation, and 100% success with 3–5 additional SPICE simulations.

关键词： Training Computational modeling Transformers SPICE Tokenization Topology Computational efficiency Transistors Integrated circuit modeling Transconductance

来源：评论

学校读者我要写书评

暂无评论

HPC Application Parameter Autotuning on Edge Devices: A Bandit Learning Approach

arXiv

引用

arXiv 2025年

作者： Hossain, Abrar Badawy, Abdel-Hameed A. Islam, Mohammad A. Patki, Tapasya Ahmed, Kishwar Department of Electrical Engineering and Computer Science The University of Toledo United States Department of Electrical and Computer Engineering New Mexico State University United States Department of Computer Science and Engineering The University of Texas Arlington United States Center for Applied Scientific Computing Lawrence Livermore National Laboratory

The growing necessity for enhanced processing capabilities in edge devices with limited resources has led us to develop effective methods for improving high-performance computing (HPC) applications. In this paper, we introduce LASP (Lightweight Autotuning of Scientific Application Parameters), a novel strategy designed to address the parameter search space challenge in edge devices. Our strategy employs a multi-armed bandit (MAB) technique focused on online exploration and exploitation. Notably, LASP takes a dynamic approach, adapting seamlessly to changing environments. We tested LASP with four HPC applications: Lulesh, Kripke, Clomp, and Hypre. Its lightweight nature makes it particularly well-suited for resource-constrained edge devices. By employing the MAB framework to efficiently navigate the search space, we achieved significant performance improvements while adhering to the stringent computational limits of edge devices. Our experimental results demonstrate the effectiveness of LASP in optimizing parameter search on edge devices. © 2025, CC BY-NC-ND.

关键词： Edge computing

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：