检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Islam, Mohammad Aminul Xing, Wangzhi Zhou, Jun Gao, Yongsheng Paliwal, Kuldip K. School of Information and Communication Technology Griffith University Australia Department of Computer Science and Mathematics Bangladesh Agricultural University Mymensingh2202 Bangladesh Institute for Integrated and Intelligent Systems Griffith University Australia

Hyperspectral object tracking has recently emerged as a topic of great interest in the remote sensing community. The hyperspectral image, with its many bands, provides a rich source of material information of an object that can be effectively used for object tracking. While most hyperspectral trackers are based on detection-based techniques, no one has yet attempted to employ YOLO for detecting and tracking the object. This is due to the presence of multiple spectral bands, the scarcity of annotated hyperspectral videos, and YOLO’s performance limitation in managing occlusions, and distinguishing object in cluttered backgrounds. Therefore, in this paper, we propose a novel framework called Hy-Tracker, which aims to bridge the gap between hyperspectral data and state-of-the-art object detection methods to leverage the strengths of YOLOv7 for object tracking in hyperspectral videos. Hy-Tracker not only introduces YOLOv7 but also innovatively incorporates a refined tracking module on top of YOLOv7. The tracker refines the initial detections produced by YOLOv7, leading to improved object-tracking performance. Furthermore, we incorporate Kalman-Filter into the tracker, which addresses the challenges posed by scale variation and occlusion. The experimental results on hyperspectral benchmark datasets demonstrate the effectiveness of Hy-Tracker in accurately tracking objects across frames. Copyright © 2023, The Authors. All rights reserved.

关键词： Remote sensing

来源：评论

学校读者我要写书评

暂无评论

LMS-Net: A Learned Mumford-Shah Network For Few-Shot Medical Image Segmentation

arXiv

引用

arXiv 2025年

作者： Zhang, Shengdong Jia, Fan Li, Xiang Zhang, Hao Shi, Jun Ma, Liyan Ying, Shihui Department of Mathematics School of Science Shanghai University Shanghai200444 China Institute University of Utah Salt Lake CityUT84102 United States School of Computer Science and Technology East China Normal University Shanghai China School of Communication and Information Engineering Shanghai University Shanghai200444 China School of Computer Engineering and Science Shanghai University Shanghai200444 China School of Mechatronic Engineering and Automation Shanghai Key Laboratory of Intelligent Manufacturing and Robotics Shanghai University Shanghai200444 China Shanghai Institute of Applied Mathematics and Mechanics The School of Mechanics and Engineering Science Shanghai University Shanghai200072 China

Few-shot semantic segmentation (FSS) methods have shown great promise in handling data-scarce scenarios, particularly in medical image segmentation tasks. However, most existing FSS architectures lack sufficient interpretability and fail to fully incorporate the underlying physical structures of semantic regions. To address these issues, in this paper, we propose a novel deep unfolding network, called the Learned Mumford-Shah Network (LMSNet), for the FSS task. Specifically, motivated by the effectiveness of pixel-to-prototype comparison in prototypical FSS methods and the capability of deep priors to model complex spatial structures, we leverage our learned Mumford-Shah model (LMS model) as a mathematical foundation to integrate these insights into a unified framework. By reformulating the LMS model into prototype update and mask update tasks, we propose an alternating optimization algorithm to solve it efficiently. Further, the iterative steps of this algorithm are unfolded into corresponding network modules, resulting in LMS-Net with clear interpretability. Comprehensive experiments on three publicly available medical segmentation datasets verify the effectiveness of our method, demonstrating superior accuracy and robustness in handling complex structures and adapting to challenging segmentation scenarios. These results highlight the potential of LMS-Net to advance FSS in medical imaging applications. Our code will be available at: https:// ***/SDZhang01/LMSNet Copyright © 2025, The Authors. All rights reserved.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Mathematical Modeling of Signal Processing Using Discrete Wavelet Transform and Non-Positional Modular Code

Mathematical Modeling of Signal Processing Using Discrete Wa...

引用

International Russian Automation Conference (RusAutoCon)

作者： Igor A. Kalmykov Alersandr A. Olenev Nikita K. Chistousov Department of Information Security of Automated Systems North-Caucasus Federal University Stavropol Russia Department of Mathematics Computer Science and Digital Educational Technologies Stavropol State Pedagogical Institute Stavropol Russia

Currently, the orthogonal frequency multiplexing (OFDM) method is the basis for the construction of most wireless communication systems. This is due to the fact that, due to the orthogonality of the carriers, OFDM systems have a high information transfer rate. The achievement of high speed was due to the use of a number of technical solutions. Firstly, the developers proposed to reduce the length of cyclic prefixes. Secondly, they began to increase the QAM constellation. Thirdly, due to the reduction of the correcting abilities of the codes, the coding speed was increased. Fourth, they began to use spatial coding (MIMO) methods. At the same time, signal processing based on forward and inverse fast Fourier transforms (FFT) was not modified. The paper considers a mathematical model of signal processing using discrete wavelet transform (DWT) and non-positional arithmetic code. The implementation of this model will lead to an increase in the transmission rate in OFDM systems. Firstly, DWT allows you to perform digital signal processing in less time compared to FFT. Secondly, the use of modular residue class codes (MCRS) provides a parallel organization of modular operations, which include addition, subtraction and multiplication. Therefore, mathematical modeling of the signal processing process using discrete wavelet transform and non-positional arithmetic code is an urgent task.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A multi-objective and CFD based optimization of roof-flap geometry and position for simultaneous drag and lift reduction

引用

Propulsion and Power Research 2024年第1期13卷 26-45页

作者： Mohammad Rostamzadeh-Renani Mohammadreza Baghoolizadeh S.Mohammad Sajadi Reza Rostamzadeh-Renani Narjes Khabazian Azarkhavarani Soheil Salahshour Davood Toghraie Energy Department Politecnico di MilanoVia Lambruschini 4Milan 20156Italy Department of Mechanical Engineering Shahrekord UniversityShahrekord 88186-34141Iran Department of Nutrition Cihan University-ErbilKurdistan RegionIraq Department of Mechanical Engineering Jami Institute of TechnologyIsfahanIran Faculty of Engineering and Natural Sciences Istanbul Okan UniversityIstanbulTurkey Faculty of Engineering and Natural Sciences Bahcesehir UniversityIstanbulTurkey Department of Computer Science and Mathematics Lebanese American UniversityBeirutLebanon Department of Mechanical Engineering Khomeinishahr BranchIslamic Azad UniversityKhomeinishahrIran

As the transport sector is responsible for the consumption of a vast proportion of the oil produced,it is mandatory to research feasible solutions to tackle this *** appli-cation of aerodynamic attachments for passiveflow control and reducing resisting aerodynamic forces such as drag and lift is one of the most practicable ways to minimize vehicle energy *** are one of the most innovative aerodynamic attachments that can enhance theflow motion in the boundary layer at the trailing edge of the *** the present paper,theflap is designed and modeled for controlling the airflow at the roof-end of a 2D Ahmed body model,inspired by the schematic of theflap at the trailing edge of the *** a result,theflap’s geometry and position from the roof-end of the car model are parameterized,which leads to having four design *** objective functions of the present study are the vehicle’s drag coefficient and lift coefficient.25 Design of Experiment(DOE)points are considered enabling the Box-Behnken ***,each DOE point is modeled in the computational domain,and theflow-field around the model is simulated using Ansys Fluent *** results obtained for the DOE points are employed by different regressors,and the relation between design variables and objective functions is extracted using *** GMDH-ANN is then coupled with three types of optimization algorithms,among which the Genetic algorithm proves to have the most ideal coupling process for ***,af-ter analyzing the variations in the geometry and position of the roofflap from the car roof-end,the roof-flap with specifications of L=0.1726 m,a=5.0875°,H=0.0188 m,and d=0.241 m can optimize the car drag and lift coefficients by 21.27% and 19.91%,*** present research discusses the opportunities and challenges of optimal design roof-flap geometry and its influence on car aerodynamic performance.

关键词： Drag coefficient Lift coefficient Roofflap Computationalfluid dynamics Artificial neural

来源：评论

学校读者我要写书评

暂无评论

Efficient Pooling Designs and Screening Performance in Group Testing for Two Types of Defectives

Efficient Pooling Designs and Screening Performance in Group...

引用

IEEE Annual International computer Software and Applications Conference (COMPSAC)

作者： Hiroyasu Matsushima Yusuke Tajima Xiao-Nan Lu Masakazu Jimbo Data Science and AI Innovation Reseach Promotion Center Shiga University Hikone Japan Department of Electrical Electronic and Computer Engineering Gifu University Gifu Japan Center for Training Professors in Statistics The Institute of Statistical Mathematics Tokyo Japan

ISBN: (数字)9798350376968

ISBN: (纸本)9798350376975

Group testing is utilized in the case when we want to find a few defectives among a large amount of items. Testing $n$ items one by one requires $n$ tests, but if the ratio of defectives is small, group testing is an efficient way to reduce the number of tests. Many researches have been developed for group testing for a single type of defectives. In this paper, we consider the case where two types of defective A and B exist. For two types of defectives, we develop a belief propagation algorithm to compute the marginal posterior probability of defectives. Furthermore, we construct several kinds of collections of pools in order to test for A and B. And, by utilizing our belief propagation algorithm, we evaluate the performance of group testing by conducting simulations.

关键词： Geometry Computational modeling Posterior probability Software algorithms Software Testing Belief propagation

来源：评论

学校读者我要写书评

暂无评论

Guest Editorial: Sentiment Analysis as a Multidisciplinary Research Area

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2022年第5期3卷 638-641页

作者： Cambria, Erik Xing, Frank Thelwall, Mike Welsch, Roy Nanyang Technological University School of Computer Science and Engineering Singapore639798 Singapore National University of Singapore Department of Information Systems and Analytics School of Computing 117418 Singapore University of Wolverhampton School of Mathematics and Computing WolverhamptonWV1 1LY United Kingdom Massachusetts Institute of Technology Sloan School of Management CambridgeMA02139 United States

The papers in this special section focus on bringing multidisciplinary knowledge into sentiment analysis. The last two decades have witnessed an enormous amount of research works on sentiment analysis and significant progress has been made. In terms of depth, finer-grained semantic schemas are defined, such as aspect, target, and category. In terms of broadness, multimodal sentiment analysis touches audio and video channels. Conversational sentiment analysis considers contextual and time-dependent relations and domain-specific sentiment analysis settings tackles real-life challenges, including but not limited to user profiling, financial prediction, abusive language detection, and mental health.

关键词： Special Issues And Sections Sentiment Analysis Computational Modeling Semantics Neural Networks Market Research Recommender Systems Knowledge Based Systems Fuzzy Logic

来源：评论

学校读者我要写书评

暂无评论

Robustifying State-space Models for Long Sequences via Approximate Diagonalization

arXiv

引用

arXiv 2023年

作者： Yu, Annan Nigmetov, Arnur Morozov, Dmitriy Mahoney, Michael W. Erichson, N. Benjamin Center for Applied Mathematics Cornell University United States Lawrence Berkeley National Laboratory United States International Computer Science Institute Department of Statistics University of California at Berkeley United States

State-space models (SSMs) have recently emerged as a framework for learning long-range sequence tasks. An example is the structured state-space sequence (S4) layer, which uses the diagonal-plus-low-rank structure of the HiPPO initialization framework. However, the complicated structure of the S4 layer poses challenges;and, in an effort to address these challenges, models such as S4D and S5 have considered a purely diagonal structure. This choice simplifies the implementation, improves computational efficiency, and allows channel communication. However, diagonalizing the HiPPO framework is itself an ill-posed problem. In this paper, we propose a general solution for this and related ill-posed diagonalization problems in machine learning. We introduce a generic, backward-stable "perturb-then-diagonalize" (PTD) methodology, which is based on the pseudospectral theory of non-normal operators, and which may be interpreted as the approximate diagonalization of the non-normal matrices defining SSMs. Based on this, we introduce the S4-PTD and S5-PTD models. Through theoretical analysis of the transfer functions of different initialization schemes, we demonstrate that the S4-PTD/S5-PTD initialization strongly converges to the HiPPO framework, while the S4D/S5 initialization only achieves weak convergences. As a result, our new models show resilience to Fourier-mode noise-perturbed inputs, a crucial property not achieved by the S4D/S5 models. In addition to improved robustness, our S5-PTD model averages 87.6% accuracy on the Long-Range Arena benchmark, demonstrating that the PTD methodology helps to improve the accuracy of deep learning models. Copyright © 2023, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

NEW ITERATIVE ALGORITHMS FOR ESTIMATION OF ITEM FUNCTIONING

arXiv

引用

arXiv 2023年

作者： Hladká, Adéla Martinková, Patrícia Brabec, Marek Institute of Computer Science The Czech Academy of Sciences Prague Czech Republic Faculty of Mathematics and Physics Charles University Prague Czech Republic Faculty of Education Charles University Prague Czech Republic

This paper explores innovations to parameter estimation in generalized linear and nonlinear models, which may be used in item response modeling to account for guessing/pretending or slipping/dissimulation and for the effect of covariates. We introduce a new implementation of the EM algorithm and propose a new algorithm based on the parametrized link function. The two novel iterative algorithms are compared to existing methods in a simulation study. Additionally, the study examines software implementation, including the specification of initial values for numerical algorithms and asymptotic properties with an estimation of standard errors. Overall, the newly proposed algorithm based on the parametrized link function outperforms other procedures, especially for small sample sizes. Moreover, the newly implemented EM algorithm provides additional information regarding respondents’ inclination to guess or pretend and slip or dissimulate when answering the item. The study also discusses applications of the methods in the context of the detection of differential item functioning and addresses the measurement error. Methods are offered in the difNLR package and in the interactive application of the ShinyItemAnalysis package;demonstration is provided using real data from psychological and educational assessments. © 2023, CC BY.

关键词： Nonlinear simulations

来源：评论

学校读者我要写书评

暂无评论

The complexity of contracts

The complexity of contracts

引用

作者： Dütting, P. Roughgarden, T. Talgam-Cohen, I. Department of Mathematics London School of Economics Houghton Street LondonWC2A 2AE United Kingdom Department of Computer Science Columbia University New YorkNY10027 United States Department of Computer Science Technion Israel Institute of Technology Haifa3200003 Israel

We initiate the study of computing (near-)optimal contracts in succinctly representable principal-agent settings. Here optimality means maximizing the principal's expected payoff over all incentive-compatible contracts-known in economics as "second-best"solutions. We also study a natural relaxation to approximately incentive-compatible contracts. We focus on principalagent settings with succinctly described (and exponentially large) outcome spaces. We show that the computational complexity of computing a near-optimal contract depends fundamentally on the number of agent actions. For settings with a constant number of actions, we present a fully polynomialtime approximation scheme (FPTAS) for the separation oracle of the dual of the problem of minimizing the principal's payment to the agent, and we use this subroutine to efficiently compute a δ-incentive-compatible (δ-IC) contract whose expected payoff matches or surpasses that of the optimal IC contract. With an arbitrary number of actions, we prove that the problem is hard to approximate within any constant c. This inapproximability result holds even for δ-IC contracts where δ is a sufficiently rapidly-decaying function of c. On the positive side, we show that simple linear δ-IC contracts with constant δ are sufficient to achieve a constant-factor approximation of the "first-best"(full-welfare-extracting) solution, and that such a contract can be computed in polynomial time. © 2021 Society for Industrial and Applied mathematics.

关键词： Computational complexity

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Approaches for Intrusion Detection with Emerging Cybersecurity Challenges

Deep Learning Approaches for Intrusion Detection with Emergi...

引用

2023 International Conference on Sustainable Communication Networks and Application, ICSCNA 2023

作者： Hemalatha, S. Mahalakshmi, M. Vignesh, V. Geethalakshmi, M. Balasubramanian, D. Jose, Anand A. St. Joseph's Institute of Technology Department of Electrical and Electronics Engineering Chennai600119 India Srm Institute of Science and Technology Department of Networking and Communications Kattankulathur Chennai603203 India Karpagam College of Engineering Department of Computer Science and Engineering Coimbatore641032 India Kcg College of Technology Department of Mathematics Chennai600097 India Jaya Engineering College Department of Mca Chennai602024 India Kcg College of Technology Department of Ece Tamil Nadu Chennai600097 India

ISBN: (纸本)9798350313987

Intrusion Detection Systems (IDS) play a pivotal role in preventing computer networks from an ever-evolving landscape of cybersecurity threats. As cyberattacks become increasingly sophisticated and diversified, the need for robust and adaptable intrusion detection models has never been more critical. This study explores the application of Deep Neural Networks (DNNs), Convolutional Neural Networks (CNNs), and Recurrent Neural Networks (RNNs) in the context of intrusion detection, particularly in addressing emerging cybersecurity challenges. The performance of deep learning (DL)-based IDSs is highly dependent on the dataset used and no benchmark models for intrusion detection have been found. In this work, a recent dataset CIC-DDoS-2019 is used for the detection of DDOS (distributed deny of services). This work underscores the importance of adopting DL techniques, including DNNs, CNNs, and RNNs, for intrusion detection in the face of emerging cybersecurity challenges. It highlights the potential of these neural network architectures to enhance network security, adapt to evolving threats, and mitigate the risks posed by modern cyber adversaries. Moreover, it encourages further research and development in this field to stay ahead of the ever-changing cybersecurity landscape. © 2023 IEEE.

关键词： DDoS attacks Deep learning Distributed denial-of-service (DDoS) Intrusion detection system Machine Learning

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：