检索结果-内蒙古大学图书馆

IEEE International Symposium on Information Theory

作者： Aristomenis Tsopelakos Olgica Milenkovic Department of Electrical and Computer Engineering Coordinated Science Laboratory University of Illinois at Urbana-Champaign

ISBN: (数字)9798350382846

ISBN: (纸本)9798350382853

In our problem, we are given access to a number of sequences of nonnegative i.i.d. random variables, whose realizations are observed sequentially. All sequences are of the same finite length. The goal is to pick one element from each sequence in order to maximize a reward equal to the expected value of the sum of the selections from all sequences. The decision on which element to pick is irrevocable, i.e., rejected observations cannot be revisited. Furthermore, the procedure terminates upon having a single selection from each sequence. Our observation constraint is that we cannot observe the current realization of all sequences at each time instant. Instead, we can observe only a smaller, yet arbitrary, subset of them. Thus, together with a stopping rule that determines whether we choose or reject the sample, the solution requires a sampling rule that determines which sequence to observe at each instant. The problem can be solved via dynamic programming, but with an exponential complexity in the length of the sequences. In order to make the solution computationally tractable, we introduce a decoupling approach and determine each stopping time using either a single-sequence dynamic programming, or a Prophet Inequality inspired threshold method, with polynomial complexity in the length of the sequences. We prove that the decoupling approach guarantees at least 0.745 of the optimal expected reward of the joint problem. In addition, we describe how to efficiently compute the optimal number of samples for each sequence, and its' dependence on the variances.

关键词： Polynomials Finite element analysis Dynamic programming Complexity theory Random variables Information theory

来源：评论

学校读者我要写书评

暂无评论

Adaptive Fine-Grained Feature Mining and RoI Feature Interaction Network for Small Object Detection in Aerial Images

Adaptive Fine-Grained Feature Mining and RoI Feature Interac...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Fangmin Xie Zhan Xiong Zhuoyue Wang Kun Yang Guoqiang Xiao College of Computer and Information Science Southwest University China Institute of Advanced Computer Science Leiden University Netherlands Department of Electrical Engineering and Computer Sciences University of California Berkeley Berkeley USA

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Object detection in drone view images remains challenging due to small-scale objects distributed non-uniformly and exhibiting weak semantic features. Additionally, external factors such as lighting conditions, viewing angles, and background clutter further complicate the detection of small objects. To address these issues, we propose a small object detector called Adaptive Fine-grained Feature Mining Network (AFMNet) based on an anchor-free framework. Firstly, an Adaptive Fine-grained Feature Mining (AFFM) module is designed to generate Regions of Interest (RoI) on high-resolution feature maps and mine fine-grained information that benefits the representation of small objects. Secondly, an Inter-RoI Feature Interaction (IRFI) block is developed to efficiently integrate complementary cues— namely, deep layer semantic information and shallow layer spatial details—and enhance RoI features. Experiments on the VisDrone dataset demonstrate that our method achieves state-of-the-art performance. The source code of our AFMNet is available at https://***/xiefangmin/AFMNet.

关键词： Adaptive systems Source coding Semantics Lighting Object detection Detectors Signal processing Feature extraction Speech processing Drones

来源：评论

学校读者我要写书评

暂无评论

Hierarchy-Aware and Channel-Adaptive Semantic Communication for Bandwidth-Limited Data Fusion

引用

IEEE Wireless Communications Letters 2025年第6期14卷 1748-1752页

作者： Guo, Lei Chen, Wei Sun, Yuxuan Ai, Bo Pappas, Nikolaos Quek, Tony Beijing Jiaotong University State Key Laboratory of Advanced Rail Autonomous Operation China Beijing Jiaotong University School of Electronic and Information Engineering Beijing China Linköping University Department of Computer and Information Science Linköping Sweden Singapore University of Technology and Design Information Systems Technology and Design Singapore

Obtaining high-resolution hyperspectral images (HR-HSI) is costly and data-intensive, making it necessary to fuse low-resolution hyperspectral images (LR-HSI) with high-resolution RGB images (HR-RGB) for practical applications. However, traditional fusion techniques, which integrate detailed information into the reconstruction, significantly increase bandwidth consumption compared to directly transmitting raw data. To overcome these challenges, we propose a hierarchy-aware and channel-adaptive semantic communication approach for bandwidth-limited data fusion. A hierarchical correlation module is proposed to preserve both the overall structural information and the details of the image required for super-resolution. This module efficiently combines deep semantic and shallow features from LR-HSI and HR-RGB. To further reduce bandwidth usage while preserving reconstruction quality, a channel-adaptive attention mechanism based on Transformer is proposed to dynamically integrate and transmit the deep and shallow features, enabling efficient data transmission and high-quality HR-HSI reconstruction. Experimental results on the CAVE and Washington DC Mall datasets demonstrate that our method outperforms single-source transmission, achieving up to a 2 dB improvement in peak signal-to-noise ratio (PSNR). Additionally, it reduces bandwidth consumption by two-thirds, confirming its effectiveness in bandwidth-constrained environments for HR-HSI reconstruction tasks. © 2012 IEEE.

关键词： Bandwidth

来源：评论

学校读者我要写书评

暂无评论

Practical aberration correction using deep transfer learning with limited experimental data

引用

Optics Express 2025年第6期33卷 14431-14444页

作者： Kok, Yong En Bentley, Alexander Parkes, Andrew J. Somekh, Michael G. Wright, Amanda J. Pound, Michael P. University of Nottingham School of Computer Science NottinghamNG8 1BB United Kingdom Optics and Photonics Group Department of Electrical and Electronic Engineering University of Nottingham NottinghamNG7 2RD United Kingdom Research Center for Humanoid Sensing Zhejiang Laboratory Hangzhou3111100 China

Adaptive optics is a technique for correcting aberrations and improving image quality. When adaptive optics was first used in microscopy, it was common to rely on iterative approaches to determine the aberrations present. It is advantageous to avoid iteration, and therefore there has been a shift to deep learning for aberration prediction. However, issues remain regarding the practicalities of machine learning for adaptive optics, an important one being the requirement for a large training dataset. Here, we explore transfer learning to overcome this need for data by pre-training a network on a large simulated dataset and fine-tuning it with reduced experimental data for application in an experimental setting. We demonstrate that the pre-trained network can make noticeable improvements with fine-tuning on just 24 experimental samples. To further enhance practicality, we significantly extend the range of aberrations present, predicting up to 25 Zernike modes with each coefficient ranging from −1 to 1, and perform a thorough analysis of the type and magnitude of phase-diversity required in the input data for a successful network. Our approach demonstrates substantial aberration reduction on experimental data for 10 Zernike modes, with an average 73% decrease in RMS wavefront error from 1.81 to 0.48 rad when correction is applied. This method achieves complete experimental image capture and aberration inference at rates comparable to the image acquisition time of a typical laser scanning microscope. Additionally, we consider the benefits of further improvements via an iterative step. As such, this work addresses some of the key practical hurdles that remain in the use of deep learning for aberration prediction and correction. © 2025 Optica Publishing Group (formerly OSA). All rights reserved.

关键词： Adaptive optics

来源：评论

学校读者我要写书评

暂无评论

Advancing Vascular Dementia Diagnosis Through Salp Swarm Optimized Convolutional Neural Network 8

Advancing Vascular Dementia Diagnosis Through Salp Swarm Opt...

引用

8th IEEE International Conference on Computational System and Information Technology for Sustainable Solutions, CSITSS 2024

作者： Mathkunti, Nivedita Manohara Ananthanagu, U. Abdulpur, Vidya N. Mangaiyarkarasi, D. Anitha, S. Alliance School of Advanced Computing Alliance Univerisity Department of Computer Science & Engineering Karnataka Bengaluru India Dr.D.Y Patil Pratishthan's College of Engineering Department of Electrical Engineering Maharastra Kolhapur India Institute of Science and Technology Avadi India Department of Artificial Intelligence & Machine Learning Karnataka Bengaluru562106 India

ISBN: (纸本)9798331505462

Vascular Dementia (VD) is a form of cognitive impairment seen in individuals, restricting from problems with blood vessels supplying the brain. These issues lead to strokes or reduced blood flow and gradually lead to the development of vascular dementia. Since it tends to progress slowly, diagnosis often comes after significant brain damage making prevention is crucial. Early identification of risk factors are vital to mitigate the risk of irreversible cognitive decline associated with vascular dementia. Accordingly, this research proposes an optimized neural network model for predicting vascular dementia at its earlier to improve the quality of life. The initial stage in this research involves preprocessing, where the input data is cleaned and normalized. These steps ensure that the vascular dementia dataset is cleaned of inconsistencies and prepared with normalized features for accurate analysis and modeling. Following preprocessing, relevant data features are selected using the chi-square approach, making the classification task easier. Finally, classification is performed utilizing a Convolutional Neural Network (CNN) and for fine-tuning of CNN parameters Salp Swarm Optimization (SSO) Algorithm is implemented. The execution of proposed work is processed using Python software and the outcome reveals superior performance in terms of accuracy and precision. Thereby the proposed method provides better performance with an accuracy of 94.5%, contrasted to existing methods in predicting VD. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Microstructure controlled synthesis of Ni,N-codoped CoP/carbon fiber hybrids with improving reaction kinetics for superior sodium storage

引用

Journal of Materials science & Technology 2022年第4期99卷 184-192页

作者： Huijun Li Xiaomin Wang Zhenxin Zhao Rajesh Pathak Siyue Hao Xiaoming Qiu Qiquan Qiao College of Materials Science and Engineering Taiyuan University of TechnologyTaiyuan 030024China Shanxi Key Laboratory of New Energy Materials and Devices Taiyuan University of TechnologyTaiyuan 030024China Department of Electrical Engineering and Computer Science Center for Advanced Photovoltaics and Sustainable EnergySouth Dakota State UniversityBrookingsSD 57007USA Mechanical&Aerospace Engineering Syracuse UniversitySyracuseNY 13244USA

Transition-metal phosphides(TMPs)-based hybrid structure have received considerable attention for efficient sodium storage owing to their high capacity and decent ***,the volume expansion&the poor electronic conductivity of TMPs,the poor-rate capability,and fast capacity decay greatly hinder its practical *** address these issues,a low-cost and facile strategy for the synthesis of Ni,N-codoped graphitized carbon(C)and cobalt phosphide(CoP)embedded in carbon fiber(Ni-CoP@CN⊂CF)as self-supporting anode material is demonstrated for the first *** graphitized carbon and carbon fiber improve the electrical conductivity and inhibit the volume expansion *** addition to that,the microporous structure,and ultrasmall sized Ni-CoP offer a high surface area for electrolyte wettability,short Na-ion diffusion path and fast charge transport *** a result,outstanding electrochemical performance with an average capacity decay of 0.04%cycle^(−1)at 2000 mA g^(−1),an excellent rate capability of 270 mAh g^(−1)@2000 mA g^(−1)and a high energy density of~231.1 Wh kg^(−1)is achieved with binder-free self-supporting anode *** work shows a potential for designing binder-free and high energy density sodium-ion batteries.

关键词： Ultrasmall Ni-CoP Carbon shell Microporous structure High Na^(+)diffusion coefficients Sodium-ion batteries

来源：评论

学校读者我要写书评

暂无评论

Learning Multi-scale Representations with Single-stream Network for Video Retrieval

Learning Multi-scale Representations with Single-stream Netw...

引用

2023 IEEE/CVF Conference on computer Vision and Pattern Recognition Workshops, CVPRW 2023

作者： Wang, Chia-Hui Tseng, Yu-Chee Chiang, Ting-Hui Chen, Yan-Ann National Yang Ming Chiao Tung University Department of Computer Science Taiwan National Yang Ming Chiao Tung University College of AI Taiwan Chunghwa Telecom Laboratories Advanced Technology Laboratory Taoyuan Taiwan Yuan Ze University Department of Computer Science and Engineering Taiwan

ISBN: (纸本)9798350302493

With the explosive growth of video contents in the Internet, video retrieval has become an important issue that can benefit video recommendation and copyright detection. Since the key features of a video may distribute in distant regions of a lengthy video, several works have made a success by exploiting multi-stream, multi-scale architectures to learn and merge distant features. However, a multi-stream network is costly in terms of memory and computing overhead. The number of scales and these scales are handcrafted and fixed once a model is finalized. Further, being more complicated, multi-stream networks are more prone to being overfitting and lead to poorer generalization. This paper proposes a single-stream network with built-in dilated spatial and temporal learning capability. By combining with modern techniques, including Denoising Autoencoder, Squeeze-and-Excitation Attention, and Triplet Comparative Mechanism, our model achieves state-of-the-art performance in several video retrieval tasks on the FIVR200K, CC WEB VIDEO, and EVVE datasets. © 2023 IEEE.

关键词： Excited states

来源：评论

学校读者我要写书评

暂无评论

SF_(6)decomposition and insulation condition monitoring of GIE:A review

引用

High Voltage 2021年第6期6卷 955-966页

作者： Fuping Zeng Haotian Li Hongtu Cheng Ju Tang Yilu Liu School of Electrical Engineering and Automation Wuhan UniversityWuhanChina Department of Electrical Engineering and Computer Science USA Oak Ridge National Laboratory KnoxvilleTennesseeUSA

The statistics of the Conference International des Grands Reseaux Electriques(CIGRE)indicate that the operational reliability of SF6 gas‐insulated equipment(GIE)is very high;however,the failure rate of the GIE in operation is much higher than that of the IEC standard,and the fault occurs frequently in the GIE at a high voltage *** reason is due to the complex and strong on‐site electromagnetic interference environment and fully enclosed structure of *** key method and technology for effective on‐line monitoring and fault diagnosis of GIE are still *** the partial strong electromagnetic energy and high temperature induced by early latent insulation faults in the equipment,SF6 gas insulation presents different degrees of *** decomposition products mainly include SO_(2)F_(2),SOF_(2),SO_(2),HF,and *** decomposition characteristics of SF6 are closely related to the property of insulation *** present,this area is attracting attention from the power industry and research *** study summarises the current research on SF_(6)decomposition component analysis(DCA).The content mainly includes the latest progress of SF_(6)decomposition characteristics and mechanism under fault conditions,and fault diagnosis methods based on decomposition components.

关键词： insulation fault diagnosis

来源：评论

学校读者我要写书评

暂无评论

Interactive Agriculture relying on Artificial Intelligence and Convolution Neural Network 2

Interactive Agriculture relying on Artificial Intelligence a...

引用

2nd IEEE International Conference on Intelligent and Innovative Technologies in Computing, electrical and Electronics, ICIITCEE 2024

作者： Nandhini, J. Mohanapriya, D. Kalaiarasi, G. Lakshmipriya, N. Arunprasath, M.M. Jai Shriram Engineering College Department of Electronics and Communication Engineering Tamil Nadu Tiruppur India Vignan's Foundation for Science Technology & Research Department of Advanced Computer Science and Engineering Andrapradesh Guntur India Jai Shriram Engineering College Department of Electrical and Electronics Engineering Tamil Nadu Tiruppur India Ksr Institute for Engineering and Technology Department of Electronics and Communication Engineering Tamil Nadu Tiruchengode India

ISBN: (纸本)9798350306415

Crop production is more complicated than farming, as it involves a lot of labor and very specific humid conditions in both the soil as well as the atmosphere in which to grow them. Farming has been made easier by technology, such as smart irrigation and precise plant nutrition, but these systems have a drawback: they spray water uniformly on all crops, even if some of them need little. An agricultural smart spraying system minimizes the impact on the environment while targeting chemical spray applications. In smart sprayers, which in most cases also have a targeted detection and spraying system, the targeted sensor forms the basis of precision spraying management. Despite the fact that excess water can be sprayed, it is not recommended. A crop can be protected against being affected by possible rots if moisture sensors and float sensors are used as part of the Intelligent Spraying System. Using this technology, it is possible to stop crops from being destroyed by climate change. The crop won't be harmed by pest sickness because of using Convolutional Neural Network (CNN) algorithm. This paper examines the conventional techniques for identifying plant diseases and pests. Agriculture productivity is seriously threatened by agricultural illnesses and insect pests. If pests are not found and identified quickly, they might result in considerable economic losses. So by using CNN those disease are classified in earlier stage and crops are being protected. © 2024 IEEE.

关键词： Artificial Intelligence CNN Image Processing Smart Irrigation

来源：评论

学校读者我要写书评

暂无评论

A Study of Trot Gait Control System of a Quadruped Robot Using IMU Sensor 61

A Study of Trot Gait Control System of a Quadruped Robot Usi...

引用

61st Annual Conference of the Society of Instrument and Control Engineers of Japan, SICE 2022

作者： Mori, Kimihiro Ohkubo, Tomoyuki Kobayashi, Kazuyuki Watanabe, Kajiro Tian, Kaiqiao Sebi, Nashwan J. Cheok, Ka C. Hosei University Graduate School of Science and Engineering Tokyo Japan Advanced Institute of Industrial Technology Tokyo Japan Oakland University School of Engineering & Computer Science Electrical & Computer Engineering Department RochesterMI United States

ISBN: (数字)9784907764784

ISBN: (纸本)9784907764784

This paper describes a stable trot gait control system for a quadruped robot that enables faster walking with stability. Compared to other gait types, it is possible to walk faster with relatively little energy expenditure, which means trot gait is suitable to control for the quadruped robot. However, trot gait does not prevent roll and pitch angle fluctuations, essentially, even when walking on flat terrain, because two legs temporarily support the robot body. Therefore, to suppress the effect of this fluctuation and stabilize the posture of the robot body, we proposed a trot gait control system using an IMU sensor and achieved faster walking without slipping. To confirm the validity of the proposed stable trot gait control system, we built a quadruped robot based on TurtleBot3 chassis and implement our system. To validate the system's effectiveness, we carried out several experiments running on various types of terrain, including sloping terrain. © 2022 The Society of Instrument and Control Engineers - SICE.

关键词： Control 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：