检索结果-内蒙古大学图书馆

An effective deep learning model for classifying diseases on strawberry leaves and estimating their severity based on the multi-task U‑Net

引用

Multimedia Tools and Applications 2024年 1-22页

作者： Nguyen, Dung Kim Choi, Yeong Soo Lee, Jang Hoon Tran, Minh Trieu Xin, XiangHui Department of Convergence Biosystems Engineering Chonnam National University Yongbong-ro 77 Gwangju61186 Korea Republic of Interdisciplinary Program in IT-Bio Convergence System Chonnam National University Yongbong-ro 77 Gwangju61186 Korea Republic of Agricultural Solutions BASF Korea Ltd Seoul04518 Korea Republic of Department of Artificial Intelligence Convergence Chonnam National University Yongbong- ro 77 Gwangju61186 Korea Republic of

Gray mold, powdery mildew and tip burn are common diseases on strawberries, directly or indirectly affecting annual strawberry yield. Due to the serious damage caused by these diseases, the identification to control them has attracted great attention for many years. This paper presents an effective method based on multi-task U-Net to distinguish types and estimate the severity of the diseases on strawberry leaves. A dataset of 2,130 images containing gray mold, powdery mildew, tip burn, and healthy leaves was collected. The proposed model performs three tasks including disease classification, segmentation of the infected regions, and segmentation of interested leaf regions to identify diseases and their severity. We investigated the proposed model and corresponding cascade structures with the backbones of the VGGNet, ResNet, and DenseNet on our dataset. The results demonstrate that our model achieved better performance with parameters reduced by 3 to 5 times and execution times faster by 8 to 23 times compared to the cascade structures when they performed the same tasks. Moreover, the proposed model with the VGG16 backbone has the highest effectiveness. Notably, compared to U-Net and some state-of-the-art segmentation models, the identification for the early-stage diseases has been impressively improved with mDice and mIoU metrics increasing by 7-25% and 6-23%, respectively. In addition, the segmentation performance for the infected region is also higher about by 2%. The classification on a complex background dataset reached precision, recall, F1-score, and accuracy values of 99.18%, 98.90%, 99.03%, and 98.93%, respectively, higher than the performance of some existing classification models. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Plant diseases

来源：评论

学校读者我要写书评

暂无评论

Photonic Crystals with Split Ring Unit Cells for Subwavelength Light Confinement

Photonic Crystals with Split Ring Unit Cells for Subwaveleng...

引用

CLEO: Science and Innovations, S and I 2022

作者： Arnold, Kellen P. Halimi, Sami I. Allen, Joshua A. Hu, Shuren Weiss, Sharon M. Interdisciplinary Materials Science Program Vanderbilt University NashvilleTN37235 United States Department of Electrical Engineering and Computer Science Vanderbilt University NashvilleTN37235 United States

We report a split ring photonic crystal that demonstrates an order of magnitude larger peak energy density compared to traditional photonic crystals. The split ring offers highly focused optical energy in an accessibl...

ISBN: (纸本)9781557528209

关键词：

来源：评论

学校读者我要写书评

暂无评论

Factors that Influence E-Satisfaction and Their Impact on Online Purchase Intention of Telemedicine Applications 9

Factors that Influence E-Satisfaction and Their Impact on On...

引用

9th IEEE International Conference on Computing, engineering and Design, ICCED 2023

作者： Gui, Anderes So, Idris Gautama Novantia, Dea Princes, Elfindah Sianipar, Nesti Fronika Haron, Hasnah Bina Nusantara University Information System Program School of Information Systems Jakarta Indonesia Bina Nusantara University Binus Business School Management Department Jakarta Indonesia Bina Nusantara University Binus Graduate Program-Master of Information Systems Management Information Systems Management Department Jakarta Indonesia Bina Nusantara University Faculty of Engineering Food Technology Department Jakarta Indonesia Universiti Sains Islam Malaysia Faculty of Economics and Muamalat Nilai Malaysia

ISBN: (纸本)9798350370126

This study embarked on a rigorous examination of the factors driving user satisfaction and usage behavior in the context of telehealth applications. Utilizing a well- structured Google Form survey, distributed through an array of social media platforms, including WhatsApp, Discord, Instagram, and Line, the research gathered 251 responses. The purposive sampling technique was employed to ensure that the data captured the experiences of individuals who had actively engaged with telehealth services. One hypothesis, relating to the influence of Online Reviews, was found to be unsupported. This surprising result suggests that, contrary to expectations, the opinions and reviews shared online do not significantly affect users' satisfaction or their behavioral decisions in the context of telehealth applications. This finding highlights the need for a more nuanced understanding of the factors influencing user preferences and choices in this rapidly evolving sector. Conversely, other hypotheses examining factors such as Social Influence, Facilitating Conditions, Perceived Reliability, Price Value, and Purchase Intention were substantiated. The positive influence of these factors on customer usage behavior underscores their significance in shaping user experiences and satisfaction within the telehealth ecosystem. Furthermore, the study posits that facilitating conditions should be closely aligned with technological advancements, as this alignment is conducive to the development of enhanced application features. This insight suggests that telehealth providers should continuously innovate and integrate advanced technologies to ensure that the facilitating conditions meet the ever-evolving needs and expectations of users. © 2023 IEEE.

关键词： Sales

来源：评论

学校读者我要写书评

暂无评论

Aerial multi-spectral AI-based detection system for unexploded ordnance

引用

Defence Technology（防务技术） 2023年第9期27卷 24-37页

作者： Seungwan Cho Jungmok Ma Oleg A.Yakimenko Republic of Korea Army Republic of Korea/Defense Acquisition Program Administration47Gwanmun-roGwacheon-siGyeonggi-do13809Republic of Korea Department of Defense Science Korea National Defense University1040Republic of Korea Hwangsanbul-ro Yangchon-myeonNonsan-siChungcheongnam-do33021Republic of Korea Department of Systems Engineering Naval Postgraduate SchoolUSA 1 University Circle MontereyCA93943USA

Unexploded ordnance(UXO)poses a threat to soldiers operating in mission areas,but current UXO detection systems do not necessarily provide the required safety and efficiency to protect soldiers from this *** technological advancements in artificial intelligence(AI)and small unmanned aerial systems(sUAS)present an opportunity to explore a novel concept for UXO *** new UXO detection system proposed in this study takes advantage of employing an AI-trained multi-spectral(MS)sensor on *** paper explores feasibility of AI-based UXO detection using sUAS equipped with a single(visible)spectrum(SS)or MS digital electro-optical(EO)***,it describes the design of the Deep Learning Convolutional Neural Network for UXO detection,the development of an AI-based algorithm for reliable UXO detection,and also provides a comparison of performance of the proposed system based on SS and MS sensor imagery.

关键词： Unexploded ordnance(UXO) Multispectral imaging Small unmanned aerial systems(sUAS) Object detection Deep learning convolutional neural network(DLCNN)

来源：评论

学校读者我要写书评

暂无评论

Data-Driven Incentive Mechanisms for Federated Learning in Vehicular Networks

引用

IEEE Transactions on Vehicular Technology 2025年

作者： Fathi, Faeze Montazeri, Mina N.araabi, Babak Du, Hongyang Niyato, Dusit Kebriaei, Hamed University of Tehran School of Ece College of Engineering Tehran Iran Urban Energy Systems Laboratory Dübendorf8600 Switzerland Nanyang Technological University School of Computer Science and Engineering The Energy Research Institute @ Ntu Interdisciplinary Graduate Program Singapore Nanyang Technological University School of Computer Science and Engineering Singapore School of Computer Science Tehran Iran

Centralized machine learning algorithms in vehicular networks face privacy and resource constraints. Federated Learning (FL) addresses these by enabling collaborative model training without sharing raw data. To incentivize vehicle participation despite resource constraints, we propose a contract between Roadside Units (RSUs) and vehicles, specifying required computation resources and rewards based on local data quality. The contract is designed to maximize the utility of the RSU, subject to constraints that ensure the participation of the vehicles in the task and truthful reporting of their data quality parameter. However, designing an optimal contract with classical methods is limited to specific utility functions. To extend the design methodology of optimal contracts to a broader range of utility functions, our study proposes a deep learning-based solution. Specifically, we present a novel application of Deep Neural Networks (DNNs), marking the first instance of applying DNNs in contract theory-based incentive mechanisms to achieve nearoptimal contracts. We design the DNN's structure to inherently satisfy specific constraints of the optimization problem, enhancing its efficiency and accuracy in determining the optimal contract. We utilize an augmented Lagrangian method to approximate the optimal contract in federated learning tasks. A case study on traffic sign recognition demonstrates the applicability and effectiveness of our method. Comparing our approach with the well-known generative diffusion model shows that in the FL scenarios, the DNN results are closer to the optimal contract compared to those of the generative diffusion model. © 1967-2012 IEEE.

关键词： Traffic signs

来源：评论

学校读者我要写书评

暂无评论

Analysing the Diurnal Evolution of Land Surface Temperature During Heat Wave Event at High Spatial Resolution

Analysing the Diurnal Evolution of Land Surface Temperature ...

引用

Remote Sensing Symposium (InGARSS), 2020 IEEE India Geoscience and

作者： Kukku Sara Eswar Rajasekaran Department of Civil Engineering Indian Institute of Technology Bombay Mumbai India Department of Civil Engineering Interdisciplinary Program in Climate Studies Indian Institute of Technology Bombay Mumbai India

In recent decades, frequently reported heat waves attributed to climate change have profoundly affected human health and food production and posed challenges to the sustainable development goals of a densely populated country like India. This study investigates the spatial and temporal trends of Land Surface Temperature (LST) during heat wave period, utilizing high spatio-temporal resolution LST data across two distinct geographical regions of India where instances of heat events have been reported. The high spatiotemporal LST was derived using a hybrid approach combining multi-variate spatial disaggregation and diurnal temperature cycle modelling. The diurnal parameters derived using the proposed methodology were compared during the heat wave year and a normal year. The results show that these diurnal parameters of LST can offer valuable insights into extreme temperature events and have the potential to aid in the mitigation of vegetative stress and consequences on human population during such extreme events.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Microstructural Pattern Formation during Far-from-Equilibrium Alloy Solidification

引用

Physical Review Letters 2023年第2期130卷 026203-026203页

作者： Kaihua Ji Elaheh Dorari Amy J. Clarke Alain Karma Physics Department and Center for Interdisciplinary Research on Complex Systems Northeastern University Boston Massachusetts 02115 USA Department of Metallurgical and Materials Engineering Colorado School of Mines Golden Colorado 80401 USA

We introduce a new phase-field formulation of rapid alloy solidification that quantitatively incorporates nonequilibrium effects at the solid-liquid interface over a very wide range of interface velocities. Simulations identify a new dynamical instability of dendrite tip growth driven by solute trapping at velocities approaching the absolute stability limit. They also reproduce the formation of the widely observed banded microstructures, revealing how this instability triggers transitions between dendritic and microsegregation-free solidification. Predicted band spacings agree quantitatively with observations in rapidly solidified Al-Cu thin films.

关键词： Liquid-solid phase transition Microstructure Metals Additive manufacturing Phase-field modeling Solidification

来源：评论

学校读者我要写书评

暂无评论

Model Predictive Control of Power Converters for Load Fluctuations Attenuation in DC-Microgrid With Fuel Cell and Battery Energy Storage

Model Predictive Control of Power Converters for Load Fluctu...

引用

IEEE PES Transmission and Distribution Conference and Exposition

作者： Muhammad Maaruf Muhammad Khalid Systems Engineering Department King Fahd University of Petroleum & Minerals Dhahran Saudi Arabia Interdisciplinary Research Center for Renewable Energy and Power Systems KFUPM Dhahran Saudi Arabia Electrical Engineering Department King Fahd University of Petroleum & Minerals Dhahran Saudi Arabia

ISBN: (数字)9798350316377

ISBN: (纸本)9798350316384

This paper proposes a constrained model predictive control (MPC) to attenuate the adverse impacts of load power fluctuations in a DC-microgrid incorporating fuel cells (FC) and battery energy storage systems (BES). Various constraints are imposed on the currents of the energy sources and the control inputs of the DC power converters to satisfy their practical limitations. Moreover, the DC link voltage, as well as the load voltage, is constrained within a safe range to ensure the continuous supply of power and prevent degradation of the DC-microgrid due to load power fluctuations. At each sampling instant, the constrained MPC computes a state feedback control policy that minimizes the cost function and satisfies the constraints. The MPC problem is solved using the linear matrix inequality (LMI). The simulation results demonstrate the feasibility of the proposed control scheme.

关键词： Degradation State feedback Fluctuations Simulation Fuel cells Cost function Safety

来源：评论

学校读者我要写书评

暂无评论

Characterization of Thermally Driven V -shaped MEMS Actuator Using an Alternate Nonlinear RC Model

Characterization of Thermally Driven V -shaped MEMS Actuator...

引用

International Conference on Electric Power and Energy Conversion systems (EPECS)

作者： Nidal A. Sherif Rached Dhaouadi Maher Bakri-Kassem Mechatronics Graduate Program American University of Sharjah Sharjah UAE Department of Electrical Engineering American University of Sharjah Sharjah UAE Department of Systems Design Engineering University of Waterloo Waterloo Canada

ISBN: (数字)9798331516963

ISBN: (纸本)9798331516970

This paper presents a reduced and efficient alternate model to simulate the nonlinear dynamics of a thermally driven V-shaped MEMS actuator. The experimental observation of the dynamic voltage-displacement relationship shows an overdamped response with a variable rise-time and fall-time indicating the simultaneous presence of complex energy storage and energy dissipation mechanisms. To completely characterize these mechanisms and yet have a simple representation for control, we develop an alternate model consisting of a set of ordinary nonlinear differential equations representing the behavior of a nonlinear RC circuit with variable parameters that are a function of the applied voltage. The simulation results show good agreement with the measured data and confirm the accuracy of the proposed alternate model.

关键词： Micromechanical devices Actuators Transient response Thermal variables control Energy dissipation Thermal variables measurement Mathematical models Integrated circuit modeling Voltage control Energy storage

来源：评论

学校读者我要写书评

暂无评论

Integrating Predictive Model Markup Language with Asset Administration Shell

引用

IFAC-PapersOnLine 2023年第2期56卷 9823-9830页

作者： Seung-Jun Shin Jumyung Um School of Interdisciplinary Industrial Studies Hanyang University Seoul Republic of Korea Department of Industrial & Management Systems Engineering Kyung Hee University Yongin Republic of Korea

The article presents a systematic approach to integrate Predictive Model Markup Language (PMML) with Asset Administration Shell (AAS) for manufacturing interoperability. The present system aims to exchange and share PMML, i.e., data analytics models, across AASs, i.e., asset representations of heterogeneous manufacturing assets. Furthermore, the present system is designed to automatically generate data analytics models on production machines, convert models into the PMML format, create AAS instances for the machines, and embed the PMML models onto the AAS instances. The article includes the design architecture, including a concept model, system architecture, information structure. An AAS client-server prototype is implemented to demonstrate the feasibility of the present system. In the prototype, a server creates and transmits the AAS that corresponds to a production machine and contains submodels associated with PMML-based energy prediction models derived by regression analysis and artificial neural network. A client receives and parses the AAS and its PMML models to predict energy consumed in the machine.

关键词： Asset Administration Shell Predictive Model Markup Language Interoperability Smart Factory Manufacturing Intelligence Energy Prediction

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：