检索结果-内蒙古大学图书馆

2nd International Conference on Data Science and Network engineering, ICDSNE 2024

作者： Priyadharshini Jayadurga, N. Chandralekha, M. Saleem, Kashif Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Chennai India Department of Computer Sciences and Engineering College of Applied Studies and Community Service King Saud University Riyadh Saudi Arabia

ISBN: (纸本)9789819783359

An innovative pipeline integrating wavelet transform, k-nearest neighbors (k-NN), and temporal wrapping approaches is presented in this work for the classification of eye blinks in electroencephalogram (EEG) recordings. The proposed process commences with raw EEG data preprocessing followed by wavelet treatment to extract significant time and frequency domain properties. These characteristics are subsequently utilized by a k-NN classifier to enhance the accuracy of eye blink recognition by using spatial relationships within EEG signals. Notably, the integration of a novel time wrapping technique deals with temporal dynamics, supplying the model with resilience to fluctuations in eye blink patterns across time. The pipeline’s enhanced efficiency can be observed through validation using a customized EEG dataset, particularly when navigating the time complexity associated with eye blink events. Evaluations performed against current techniques demonstrate the efficacy of the proposed solution. The outcomes not only demonstrate the methodology’s accuracy, but also its possible use in real-world settings featuring clinical diagnostics and human-computer interactions. This work conveys a strong basis for the discipline of EEG-based eye blink classification, having implications to enhance neuroscience and facilitating beneficial uses in healthcare and assistive technologies. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

Intelligent control algorithms for industrial automation systems

Intelligent control algorithms for industrial automation sys...

引用

2025 IEEE International Students' Conference on electrical, Electronics and computer Science, SCEECS 2025

作者： Sarathkumar, D. Anil, A.R. Sivadasan, J. Jayakumar, T. Manivel, Murugesan Anandhakumar, C. Engineering Kongu Engineering College Department of Electrical and Electronics Tamil Nadu 638 060 India Sree Buddha College of Engineering Pattoor Department of Computer Science &engineering Kerala Alappuzha India Psr Engineering College Department of Electronics and Communication Engineering Tamil Nadu Sivakasi626140 India Nandha Engineering College Department of Electrical and Electronics Engineering Tamil Nadu 638052 India Karpagam Institute of Technology Department of Electrical and Electronics Engineering Tamilnadu Coimbatore641021 India Sri Ramakrishna Institute of Technology Department of Electrical and Electronic Engineering Tamil Nadu Coimbatore641010 India

ISBN: (纸本)9798331529833

Industrial automation has become a cornerstone of modern manufacturing, enhancing efficiency, reliability, and scalability. The integration of intelligent control algorithms, such as fuzzy logic, neural networks, genetic algorithms, and model predictive control, has revolutionized traditional automation systems, enabling more adaptive, efficient, and robust operations. This paper explores the application and impact of these intelligent control techniques in industrial settings. It highlights how machine learning algorithms can predict equipment failures, optimize processes, and enhance decision-making in real-time. Case studies are presented to illustrate the benefits and challenges of deploying these algorithms across various automated systems, such as robotic assembly lines, process control in chemical plants, and smart warehousing. Furthermore, the paper discusses the integration of these algorithms with industrial IoT (IIoT) platforms, facilitating seamless communication and data exchange between machines and control systems. The research also addresses the challenges of implementing intelligent control, such as computational complexity, security concerns, and the need for real-time processing. Finally, this study evaluates future trends, emphasizing the potential of combining artificial intelligence and deep learning techniques with industrial automation for a more adaptive and autonomous manufacturing ecosystem. © 2025 IEEE.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Long Distance Ad-Hoc MAC Protocol for UAV Networks

Long Distance Ad-Hoc MAC Protocol for UAV Networks

引用

2025 IEEE SoutheastCon, SoutheastCon 2025

作者： Chakravarty, Sumit Banerjee, Madhushri Syed, Tamseel Mahmood Kennesaw State University Department of Electrical and Computer Engineering United States Georgia Gwinnett College Department of Information Technology United States University of Alabama in Huntsville Department of Electrical and Computer Engineering United States

ISBN: (纸本)9798331504847

Unmanned Aerial Vehicle (UAV) networks are increasingly employed in mission-critical applications, demanding efficient and reliable communication protocols to address key challenges such as energy efficiency, collision avoidance, scalability, and Quality of Service (QoS). This paper presents the development of a Long-Distance Ad-Hoc Medium Access Control (MAC) protocol tailored for UAV networks to address such issues. This novel protocol combines hybrid MAC techniques, utilizing Time Division Multiple Access (TDMA) for inter- cluster communication and Code Division Multiple Access (CDMA) for intra-cluster transmissions ensuring resource utility efficiency and high throughput. Key features of the protocol include dynamic power control, proximity-based slot allocation, UAV trajectory sharing, and priority-based scheduling to support real-time data transmissions. Simulation results demonstrate that the protocol has up to 15-20% improved energy efficiency, ensures 100% successful collision resolution, provides scalable solutions for up to 100 nodes, and enhances QoS with a success rate of up to 95% for high-priority traffic. These results highlight the protocol's effectiveness in meeting the demanding requirements of real time communications in UAV networks with potential to support large-scale, energy- efficient, and collision-free communication in dynamic environments. © 2025 IEEE.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Rural Broadband Access via Clustered Collaborative Communication

Rural Broadband Access via Clustered Collaborative Communica...

引用

2025 IEEE SoutheastCon, SoutheastCon 2025

ISBN: (纸本)9798331504847

Traditional broadband solutions are often economically and technically infeasible in rural regions due to sparse populations, high infrastructure costs, and energy inefficiencies. This makes rural broadband access a critical challenge, particularly in developing countries where socio-economic disparities and geographical barriers exacerbate the digital divide. This paper proposes an innovative cluster-based network architecture to address such challenges by leveraging low-power Customer Premise Equipment (CPE) and underutilized television white space (TVWS) frequencies. The proposed system forms collaborative clusters of CPEs that use distributed beamforming (DBF) for transmitting data to a central Point of Presence (PoP) Base Station. This approach reduces the need for extensive infrastructure and high-power transmissions, offering a cost-effective and energy-efficient solution. A two-tier uplink access protocol is introduced to optimize data transmission, ensuring scalability and reliability as rural populations grow. By addressing the unique challenges of rural connectivity, this study provides a viable pathway for expanding internet access, enabling socio-economic growth, and bridging the digital divide in underserved communities. Key contributions include dynamic clustering algorithms, scalability enhancements, and a comprehensive performance evaluation, offering a transformative model for rural broadband deployment. Simulation results demonstrate that the proposed architecture is up to 20% more energy efficient with significantly improved throughput compared to conventional rural broadband access schemes. © 2025 IEEE.

关键词： collaborative clustered CPEs distributed beamforming (DBF) underutilized television white space (TVWS)

来源：评论

学校读者我要写书评

暂无评论

3D imaging of complex specular surfaces by fusing polarimetric and deflectometric information

引用

Optica 2025年第4期12卷 446-450页

作者： Wang, Jiazhang Cossairt, Oliver Willomitzer, Florian Wyant College of Optical Sciences University of Arizona TucsonAZ85721 United States Department of Electrical and Computer Engineering Northwestern University EvanstonIL60208 United States

Accurate and fast 3D imaging of specular surfaces still poses major challenges for state-of-the-art optical measurement principles. Frequently used methods, such as phase-measuring deflectometry (PMD) or shape-from-polarization (SfP), rely on strong assumptions about the measured objects, limiting their generalizability in broader application areas such as medical imaging, industrial inspection, virtual reality, or cultural heritage analysis. In this paper, we introduce a measurement principle that utilizes a technique to effectively encode and decode the information contained in a light field reflected off a specular surface. We combine polarization cues from SfP with geometric information obtained from PMD to resolve all arising ambiguities in the 3D measurement. Moreover, our approach avoids the unrealistic orthographic imaging assumption for SfP, which significantly improves the respective results. We showcase our technique by demonstrating single-shot and multi-shot measurements on complex-shaped specular surfaces, displaying an evaluated accuracy of surface normals below 0.56◦ © 2025 Optica Publishing Group.

关键词： Medical imaging

来源：评论

学校读者我要写书评

暂无评论

Enhancing energy management and power quality in grid-connected photovoltaic systems through ESS and STATCOM integration

引用

International Journal of Advanced Manufacturing Technology 2025年 1-16页

作者： Velukutty, Kannan Nithin Kariyankattil Sundararajan, Deepa Krishnappa, Muthuraman Ramaiah Angappan, Senthil Kumar Premalatha, Munuswamy Sivaraju, Selligoundanur Subramaniyam Usha, Subramaniam SCMS School of Engineering and Technology Kerala Karukutty India Department of Physics Government College of Engineering Tamilnadu Salem India Sri Ramakrishna Engineering College Tamil Nadu Vattamalaipalayam N.G.G.O Colony Post Coimbatore641 022 India Computer Science and Engineering School of Engineering Dayananda Sagar University Bangalore India Department of Mathematics Vel Tech Rangarajan Dr. Sagunthala R & D Institute of Science and Technology Tamil Nadu Chennai India Department of Electrical and Electronics Engineering RVS College of Engineering and Technology Kannampalayam Tamil Nadu Coimbatore India Department of Electrical and Electronics Engineering Kongu Engineering College Tamilnadu Perundurai India

This paper presents a hybrid system that integrates a photovoltaic (PV) array, an energy storage system (ESS), and a Static Synchronous Compensator (STATCOM), utilizing a Quasi-Z Source Inverter (qZSI) to improve the efficiency of grid-connected power systems. The qZSI facilitates both voltage boosting and inversion in a single stage, enabling efficient power conversion from the PV array and ESS while minimizing component count. The STATCOM plays a crucial role in dynamic reactive power control and voltage regulation, while the ESS provides rapid charge and discharge capabilities, ensuring grid stability during voltage fluctuations and varying load demands. To enhance ESS battery safety, an energy management strategy is proposed, which regulates power flow to prevent overcharging and over-discharging, thereby extending its life and optimizing performance. The system operates in three distinct modes: PV mode for real power generation, PV-STATCOM mode for reactive power control and power factor correction, and ESS-STATCOM mode for emergency grid support. This integrated approach enhances power quality, improves grid flexibility, and supports the sustainable integration of renewable energy. MATLAB simulations confirm the system’s capability to maintain power quality and stabilize the grid across various operational scenarios, highlighting its efficiency and cost-effectiveness for modern power grids. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Battery management systems

来源：评论

学校读者我要写书评

暂无评论

A Multi-Modal Assessment Framework for Comparison of Specialized Deep Learning and General-Purpose Large Language Models

引用

IEEE Transactions on Big Data 2025年第3期11卷 1001-1012页

作者： Nadeem, Mohammad Sohail, Shahab Saquib Madsen, Dag Oivind Alzahrani, Ahmed Abrahim Ser, Javier Del Muhammad, Khan Aligarh Muslim University Department of Computer Science Aligarh India VIT Bhopal University School of Computing Science and Engineering Bhopal-Indore Highway Kothrikalan Madhya Pradesh Sehore466114 India University of South-Eastern Norway Norway King Saud University Computer Science Department Community College Riyadh11437 Saudi Arabia TECNALIA Derio48160 Spain Department of Communications Engineering Bilbao48013 Spain Department of Applied Artificial Intelligence School of Convergence College of Computing and Informatics Seoul03063 Korea Republic of

Recent years have witnessed tremendous advancements in Al tools (e.g., ChatGPT, GPT 4, and Bard), driven by the growing power, reasoning, and efficiency of Large Language Models (LLMs). LLMs have been shown to excel in tasks ranging from poem writing and coding to essay generation and puzzle solving. Despite their proficiency in general queries, specialized tasks such as metaphor understanding and fake news detection often require finely tuned models, posing a comparison challenge with specialized Deep Learning (DL). We propose an assessment framework to compare task-specific intelligence with general-purpose LLMs on suicide and depression tendency identification. For this purpose, we trained two DL models on a suicide and depression detection dataset, followed by testing their performance on a test set. Afterward, the same test dataset is used to evaluate the performance of four LLMs (GPT 3.5, GPT 4, Google Bard, and MS Bing) using four classification metrics. The BERT-based DL model performed the best among all, with a testing accuracy of 94.61%, while GPT 4 was the runner-up with accuracy 92.5%. Results demonstrate that LLMs do not outperform the specialized DL models but are able to achieve comparable performance, making them a decent option for downstream tasks without specialized training. However, LLMs outperformed specialized models on the reduced dataset. © 2015 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Edge detection using the Prewitt operator with fractional order telegraph partial differential equations (PreFOTPDE)

引用

Multimedia Tools and Applications 2025年第13期84卷 12329-12345页

作者： Tenekeci, Mehmet Emin Abdulazeez, Sadeq Taha Karadağ, Kerim Modanli, Mahmut Department of Computer Engineering Faculty of Engineering Harran University Şanlıurfa Turkey Department of Mathematics College of Basic Education University of Duhok Duhok Iraq Department of Electrical and Electronics Engineering Faculty of Engineering Harran University Şanlıurfa Turkey Depertmants of Mathematics Faculty of Arts and Sciences Harran University Sanliurfa Turkey

Detecting edges in image processing is an important process in image analysis or enhancement. Many methods detected edge information based on the differences in brightness values. Prewitt, the most widely used edge detection method, performs the extraction of edges by partial differential calculation. This study performed fractional order partial differential calculation with the Telegraph equation in the Prewitt method calculation. Filter coefficients are determined by using Grumwald-Letnikov series expansion for fractional partial differential equation calculation. To determine the most appropriate fraction value, five different coefficients were tried. The proposed method is compared with the conventional edge detection methods Sobel, Canny and standard Prewitt. The obtained result showed that the proposed method detects the edges more accurately. In addition, it has been compared with Prewitt fractional order edge detection (FOED) and edge detection method based on ant colony optimization (ACO) methods developed in recent years. Pratt's figure of merit (FOM) metric was used to compare these studies. It was realized that the FOM values calculated with the proposed method were more accurate than the others methods. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Edge detection

来源：评论

学校读者我要写书评

暂无评论

AI-Inspired Algorithm for the Automated Recognition of the World’s Oldest Script “Brahmi”

引用

SN computer Science 2025年第1期6卷 1-12页

作者： Jain, Trang Jain, Arpit Dwivedi, Rakesh Kumar Saxena, Rakhi College of Computing Sciences and Information Technology Teerthanker Mahaveer University UP Moradabad India Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Vaddeswaram Vijayawada India Department of Computer Science Deshbandhu College University of Delhi New Delhi India

One of the internationally known oldest script is Brahmi whose digitisation may be helpful for the archaeologists as well as it may help in the digitisation of other languages. Optical character recognition techniques with genetic algorithms will be quite useful for this purpose. In this paper, a review of the work, done on the images of Brahmi letters using CNN till now, is presented. It will be useful to check the possibility of improvement in the performance of a model used for character recognition. A model is designed where tuning of hyper parameters is done using genetic algorithms which improves the accuracy of recognition to 97%. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Ashok shilalekh Brahmi script CNN Kharavela shilalekh OCR

来源：评论

学校读者我要写书评

暂无评论

Fast Adversarial Training with Dynamic Batch-Level Attack Control 23

Fast Adversarial Training with Dynamic Batch-Level Attack Co...

引用

Proceedings of the 60th Annual ACM/IEEE Design Automation Conference

作者： Jaewon Jung Jaeyong Song Hongsun Jang Hyeyoon Lee Kanghyun Choi Noseong Park Jinho Lee College of Computing Yonsei University Department of Electrical and Computer Engineering Seoul National University

ISBN: (纸本)9798350323481

Despite the fact that adversarial training provides an effective protection against adversarial attacks, it suffers from a huge computational overhead. To mitigate the overhead, we propose DBAC, a fast adversarial training with dynamic batch-level attack control. Based on a prior study where attack strength should gradually grow throughout the training, we control the number of samples attacked per batch for better throughput. Additionally, we collect samples from multiple batches to form a pseudo-batch and attack them simultaneously for higher GPU utilization. We implement DBAC using PyTorch to show its superior throughput with similar robust accuracy compared to the prior art.

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：