检索结果-内蒙古大学图书馆

2024 Asia Pacific Conference on Innovation in Technology, APCIT 2024

作者： Suryavanshi, Ankita Mehta, Shiva Gopal, Lisa Thapliyal, Siddhant Jain, Vishal Chitkara University Institute of Engineering and Technology Chitkara University Punjab India Graphic Era Hill University Computer Science and Engineering Uttarakhand Dehradun India Graphic Era Deemed to Be University Computer Science and Engineering Uttarakhand Dehradun India Sharda University Computer Science and Engineering Sharda School of Engineering and Technology U.P. Greater Noida India

ISBN: (纸本)9798350361537

This article proposes a thorough research study between federated learning and CNNs, where both methods are utilized for the detection and severity categorization of broccoli leaf diseases. The algorithm uses the data, the separation of four clients' enthused, to point out the severity of diseases among four categories that start from mild to critical illness. A federated learning system has used client data instead of a global model to train a learning model where privacy issues have not arisen. The study of the results revealed that the FedAvg algorithm-based federated learning method, which is precisely featured with that property, kept each individual's observation data as an origin point to create a highly effective global model. The research depended on macro-economic, micro-economic, and weighted averages to reflect the model's accomplishments at both local and international levels. For all the intensely local ratings whatsoever, the customers SB-1, SB-2, and SB-3 showed a macro average score of 92.17 and 96.49, respectively, at a swell of 96.64 and 97.1. These scores reflect that the model's accuracy and recall are improving, which is constantly witnessed among clients. Scores were calculated by burden weighting. Since the first computation of the score was 92.26, sb-1. In the long term, scores for sb-2, sb-3, and sb-4 trended upward, with sb-4 finally peaking at 97.08. After the weighted averages, the micro averages are statistically one in line;the outcome is 92.24 in sb-1 and 97.08 in sb-4. This invariably drives the model to the correct result, implying data consistency. All the local areas had a macro average score of 97.65, a micro score of 97.63, and a weighted average score of 97.63. © 2024 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Advanced Double Input Layered Neural Network for medical diagnosis by using cloud computing technology 1

Advanced Double Input Layered Neural Network for medical dia...

引用

1st IEEE International Conference on Emerging Research in Computational science, ICERCS 2023

作者： Gowri, S. Jabez, J. Nirmalrani, V. Srinivasulu, Senduru Sathyabama Institute of Science & Technology Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350359756

Advanced Double Input Layered Neural Network has the potential to revolutionize medical diagnostics by solving pressing problems. It improves diagnostic precision by providing a single, unified platform for the examination of both organized and unstructured medical data. It provides real-time decision assistance and data management by utilizing the scalable, secure, and efficient data processing capabilities made possible by cloud computing. Improved patient care, more effective therapy, and higher-quality healthcare are some areas where Advanced Double Input Layered Neural Network can make a difference. Rapid, precise, and secure data analysis;the administration of multiple data sources;and realtime decision support are a few of the difficulties inherent in medical diagnosis. This research proposes an Advanced Double Input Layered Neural Network (ADILNN) with a double input layered neural community, enabling it to examine prepared and unstructured scientific data. This novel technique improves the community's mastering functionality from various data kinds. By centralizing statistics garage, processing, and analysis on the cloud, computing sources may be extra reliably accessed whilst wanted. The network's diagnostic precision and flexibility are each advanced through way of tool gaining knowledge of strategies (MLM). Because of its adaptability, ADILNN can be utilized in diverse medical fields, which include radiology, pathology, cardiology, and genetics. It permits examine genomic information, making recovery alternatives, and analyzing x-ray photos. The technique has numerous ability makes use of in healthcare, improving prognosis accuracy in diverse settings. Simulation analysis is used to gauge ADILNN's capacity by way of gauging its diagnostic accuracy, processing speed, scalability, and records protection. These research validate ADILNN's potential to improve clinical analysis, streamline facts management, and guarantee healthcare records's safe and effectiv

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Bean Leaf Disease Detection using VGG-19, Inception, and ResNet50 Algorithms 3

Comparative Analysis of Bean Leaf Disease Detection using VG...

引用

3rd International Conference on Electrical, Electronics, Information and Communication Technologies, ICEEICT 2024

作者： Agila, R. Fernandez, F. Mary Harin Saveetha University Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350369083

This research introduces a method for detecting diseases in beans leaves using the well-known deep learning algorithms ResNet50, VGG-19, and Inception. In order to aid in crop management and disease control, it is necessary to create a reliable model for disease detection in beans leaves. A dataset including pictures of both healthy and sick beans leaves is used by the suggested method. This dataset is used to train algorithms that can identify various illnesses based on their unique characteristics. Next, a different collection of photos is used to assess the trained models' ability to reliably diagnose illnesses. Experimental findings show that all three algorithms successfully distinguish between damaged and healthy leaves. In addition, while looking at the algorithms side by side, it becomes clear that the algorithms' performance varies with the kind of sickness being diagnosed. Based on these results, it seems that the algorithm should be selected with the target ailment in mind. In conclusion, our research shows that deep learning algorithms have promise for detecting diseases in beans leaves and sheds information on the ways in which their effectiveness varies. Deep learning algorithms, disease prevention, crop management, VGG-19, Inception, and ResNet50 for beans leaf disease detection. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Edge Computational Integrated Improved Data Model for Adjustment and Restoration Strategy for Smart Grid Environment 1

Edge Computational Integrated Improved Data Model for Adjust...

引用

1st IEEE International Conference on Emerging Research in Computational science, ICERCS 2023

作者： Premalatha, J. Sarojini Abirami, G. Poornima, D. Sivakumar, Gayathri Sathyabama Institute of Science & Technology Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350359756

In the current context, integrated distribution networks of the electricity grid to provide reliability during blackouts. Self-healing systems may keep functioning and reconfigure themselves in response to external breakdowns or attacks by using the efficient strategy of restoration. An intelligent decision-making framework is necessary to ensure the system as a whole can self-reconfigure in the face of external failures and attacks. Incorrect grid modifications can pose a significant threat to the power grid's security and cause delays in network traffic and packet loss. So, in this research, Edge Computational Integrated Improved Data paradigm (EC-IDM) to deal with failures and attacks in real time. By mitigating the consequences of security worries, packet loss, latency, and overcrowded networks, edge computing has greatly increased the smart grid's capacity for efficient operation and communication. Smart Grid's posterior intelligent model (SSPIM) integration enhances the system's functionality and productivity. This research introduces an EC-IDM that can detect blackout scenarios and automatically construct a restoration strategy in real time. In EC-IDM, an enhanced hierarchical coordination mechanism and a local object optimization method are combined to optimize failures and attacks in real time over many time and space axes. Differentiating features of EC-IDM include real-time data analysis, networked simulation, smart decision support, and a wealth of other useful features. The simulation results indicate that the recommended strategy achieves best in security concerns optimizes network congestion delay, and packet loss security, increases efficiency and performance compared to the other standard ways. © 2023 IEEE.

关键词： Smart power grids

来源：评论

学校读者我要写书评

暂无评论

RC Row Houses in Nepal and their Vulnerability to Impending Earthquakes

引用

Journal of The Institution of Engineers (India): Series A 2024年第1期105卷 195-207页

作者： Devkota, Binu Dahal, Aarosh Sedhain, Bikesh Karki, Aayush Maan Pokhrel, Aashish Gyawali, Tunisha Shrestha, Kshitij C. Department of Civil Engineering Pulchowk Campus Institute of Engineering Tribhuvan University Lalitpur44600 Nepal Department of Civil Engineering College of Engineering & Computer Science The University of Texas Rio Grande Valley EdinburgTexas78539 United States School of Civil and Environmental Engineering Georgia Institute of Technology AtlantaGA30332 United States Department of Civil Engineering The University of Texas at Arlington ArlingtonTX76019 United States

In the 2015 Gorkha earthquake, several buildings in Nepal suffered heavy damages;however, those related to pounding were few. As result, row housing practices with inadequate seismic gap between buildings have been prevalent. This paper aims to compare the pounding effects of the Gorkha earthquake with that of earthquakes representative of hazard level prescribed by the national seismic design code in order to shed light on the risk of pounding damages due to potential future earthquakes in Nepal. Two reinforced concrete (RC) buildings, five storied and seven storied, were modeled and subjected to time histories of eight earthquakes—Gorkha, Kobe, El Centro, Loma Prieta, Cape Mendocino, Erzincan, Landers and Northridge Olive View matched to the response spectrum of NBC 105:2020 and the actual time history of the Gorkha earthquake. Spectral matching of the earthquakes was performed in frequency domain to preserve their frequency characteristics. The simulation of pounding was facilitated by the use of Kelvin–Voigt impact element which is capable of modeling energy dissipation during impact. The results show that the pounding forces in case of matched earthquakes were 1.3–4.4 times as great as that of Gorkha earthquake. Similarly, separation gap requirement was found to be greater for the matched earthquakes than that of Gorkha earthquake. © The Institution of Engineers (India) 2024.

关键词： Earthquakes

来源：评论

学校读者我要写书评

暂无评论

Build to Win: Electrical Hardware Opportunities for STEM Education 14

Build to Win: Electrical Hardware Opportunities for STEM Edu...

引用

14th IEEE Integrated STEM Education Conference, ISEC 2024

作者： Monagle, Daniel Ponce, Eric A. Leeb, Steven B. Saathoff, Erik K. Brooks, Dylan E. Krause, Thomas C. Green, Daisy H. Massachusetts Institute of Technology Department of Electrical Engineering and Computer Science United States University of Hawai'i Department of Electrical and Computer Engineering Manoa United States

ISBN: (纸本)9798350352801

engineering outreach and introductory courses are essential for motivating and training the next generation of capable engineers. Accessibility and portability of the infrastructure for a STEM course is critical for spreading STEM education and motivation efforts. The ubiquity of student personal computers and laptops provides a convenient platform for introducing computer programming, but the relatively cumbersome equipment deemed necessary for a hands-on electronics learning environment contributes to the lack of electronic circuit hardware exposure for many K-12 students and university undergraduates. This paper presents the relatively low-cost, versatile, electronics platform we have developed at MIT upon which our new, hands-on introductory STEM electrical engineering course has been built and taught in recent years. This paper focuses its discussion on the power supply and measurement systems we have produced that allow students and instructors to engage in hands-on electric circuit exercises and projects without the requirement for expensive laboratory infrastructure. Example demonstrations of introductory laboratory exercises using our platform are shown. © 2024 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Multi-Task Improve Domain Generalization in EEG-based Emotion Classification using Feature Fusion Learning Model 14

Multi-Task Improve Domain Generalization in EEG-based Emotio...

引用

14th International Conference on Computing Communication and Networking Technologies, ICCCNT 2023

作者： Menaka, S. Joshi, Riddhi. S. Varun, M. SRM Institute of Science and Technology Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350335095

The use of Electroencephalography (EEG) signals for emotion identification tasks is common. However, the domain shift issue may cause EEG-based emotion detection models' performance to decline when used in new domains. In this article, we present a feature fusion learning model-based multi-task learning strategy to enhance domain generalization in EEG-based emotion classification. By simultaneously learning a shared feature extractor and numerous domain-specific classifiers, we train a single model to categorize emotions across various domains. Additionally, we present a feature fusion module that combines time-domain, frequency-domain, and wavelet-domain EEG characteristics. We run trials on the DEAP and SEED EEG emotion recognition datasets to assess our suggested methodology. The experimental findings show that our suggested strategy outperforms a number of cutting-edge techniques in terms of classification accuracy, both within- and across-domains. © 2023 IEEE.

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

LiDAR Sensor for Self-Driving Cars 8

LiDAR Sensor for Self-Driving Cars

引用

8th International Conference on Communication and Electronics Systems, ICCES 2023

作者： Tummala, Manoj Pravin, A. Sathyabama Institute of Science and Technology Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350396638

Progressive technological development results in modern automotive technology such as self-driving automobiles. These require sensors to collect data about nearby objects and the surroundings in order to identify lanes and make decisions. Instead of LiDAR sensors (Light Detection and Ranging), Tesla, Waymo, and other successful businesses have employed cameras for developing autonomous vehicles. The disadvantage of camera performance changing with lighting conditions can be solved by using LiDAR. The LiDAR sensor aids in 3D mapping, allowing cars to travel in a predictable environment. The use of cameras for monitoring normal driving circumstances might be difficult because of the large range of features they include, but LiDAR contains infrared light, which is a solution for concerns such as fog, rainy weather, and varied textures. Developing sensors driven by computer vision can lead to more environmental sensing for self-driving automobiles in the future. This study discusses the use of LiDAR sensors in autonomous cars. © 2023 IEEE.

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

A Near-Optimal Air Vehicle Energy Management Algorithm Using Surrogate Model 43

A Near-Optimal Air Vehicle Energy Management Algorithm Using...

引用

43rd AIAA DATC/IEEE Digital Avionics Systems Conference, DASC 2024

作者： Hsieh, Chao-Yi Wang, Yu-Shun National Chung-Shan Institute of Science and Technology Aeronautical Systems Research Division Avionics and Electrical Engineering Department Taichung40722 Taiwan

ISBN: (纸本)9798350349610

This work aims to introduce the primarily functionalities of the fighter cruise energy management (EM) problems, including Endurance (EDR), Range (RNG), and Home (HOME);discusses the optimization methods wildly used to address this issue (dynamic programming, direct method, and indirect method). The characteristics and limitations of these methods are explained, and the combination of dimensionless analysis and iterative-based method is eventually proposed. First, the relative equation of the overall system is derived. Then the complete data is iteratively generated, to figure out the optimal solution, and constructs an optimization database. Due to the real-time requirement of the fighter's application, this work applies the least squares method (LS) to establish a three-dimensional surrogate model to replace the original database, thereby improving the overall algorithm efficiency. By using the iterative-based method, the simulation results show that the SSE of the surrogate model is 0.0187, indicating excellent approximation. Additionally, the estimation of optimal airspeed takes only 0.002129 seconds, exhibiting good real-time performance. Thus, it can effectively replace the original complex and time-consuming model. © 2024 IEEE.

关键词： Dynamic programming

来源：评论

学校读者我要写书评

暂无评论

Enhancing stencil printing in PCB production using deep learning based approach for residue classification and optimization

引用

International Journal of Information Technology (Singapore) 2025年 1-11页

作者： Jain, Swati Patel, Ruhi Desai, Manad Verma, Jai Prakash Yoon, Sang Won Department of Computer Science and Engineering Institute of Technology Nirma University Gujarat Ahmedabad India Department of Systems Science and Industrial Engineering State University of New York at Binghamton Binghamton NY United States

Surface-mount technology (SMT) is the technology used in the production of printed circuit boards (PCB) plays a vital role in PCB manufacturing for applications ranging from communication devices to medical systems. A significant challenge in the stencil printing process, a critical step in PCB production, is solder paste buildup in stencil apertures, contributing to 57%–71% of soldering defects. To address this issue, this study presents a convolutional neural network (CNN)-based framework for categorizing solder paste residue levels into three classes: low, moderate, and high. The methodology involves cropping and pre-processing images of stencil apertures, which are then fed into the CNN model for classification. To enhance model robustness and prevent overfitting, data augmentation techniques, including aperture rotation, are applied, effectively increasing the dataset size and promoting generalization. The model achieves a classification accuracy of 87% during testing, with precision and recall values provided for each residue class. This approach aims to optimize the stencil printing process by enabling accurate residue categorization and facilitating targeted cleaning cycles, ultimately reducing soldering defects and improving PCB production efficiency. © Bharati Vidyapeeth's institute of computer Applications and Management 2025.

关键词： CNN Deep learning Printed sensing board (PCB) Residue build-up Stencil printing Surface mount technology (SMT) SVM VGG16

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：