检索结果-内蒙古大学图书馆

Patterns in Heuristic Optimization Algorithms: A Comprehensive Analysis

computers, Materials & Continua 2025年第2期82卷 1493-1538页

作者： Robertas Damasevicius Center of Real Time Computer Systems Kaunas University of TechnologyKaunas44249Lithuania

Heuristic optimization algorithms have been widely used in solving complex optimization problems in various fields such as engineering,economics,and computer *** algorithms are designed to find high-quality solutions efficiently by balancing exploration of the search space and exploitation of promising *** heuristic optimization algorithms vary in their specific details,they often exhibit common patterns that are essential to their *** paper aims to analyze and explore common patterns in heuristic optimization *** a comprehensive review of the literature,we identify the patterns that are commonly observed in these algorithms,including initialization,local search,diversity maintenance,adaptation,and *** each pattern,we describe the motivation behind it,its implementation,and its impact on the search *** demonstrate the utility of our analysis,we identify these patterns in multiple heuristic optimization *** each case study,we analyze how the patterns are implemented in the algorithm and how they contribute to its *** these case studies,we show how our analysis can be used to understand the behavior of heuristic optimization algorithms and guide the design of new *** analysis reveals that patterns in heuristic optimization algorithms are essential to their *** understanding and incorporating these patterns into the design of new algorithms,researchers can develop more efficient and effective optimization algorithms.

关键词： Heuristic optimization algorithms design patterns initialization local search diversity maintenance adaptation stochasticity exploration exploitation search space metaheuristics

来源：评论

学校读者我要写书评

暂无评论

Empowering Leather Quality Assurance: Leveraging Convolutional Neural Networks for Precise Defect Detection and Classification

Empowering Leather Quality Assurance: Leveraging Convolution...

引用

International Symposium on Advanced Computing and Communication (ISACC)

作者： Logeswaran K Suresh P Savitha S Swati Sah Prasanna Kumar Kr Rajadevi R Vasugi M Sujit B Akilesh T Department of Artificial Intelligence Center of Excellence in Real-Time Analytic Kongu Engineering College Erode Tamilnaud India Department of Database Systems Vellore Institute of Technology Vellore Tamil Nadu India Department of Computer Science and Engineering K.S.R College of Engineering Tiruchengode Tamil Nadu India School of Engineering and Technology Sharda University Greater Noida Uttar Pradesh India Department of Information Technology Kongu Engineering College Erode Tamilnaud India Department of Artificial Intelligence Kongu Engineering College Erode Tamilnaud India

ISBN: (数字)9798331523893

ISBN: (纸本)9798331523909

Ensuring the development of high-quality leather products in the industrial sector requires a strong focus on leather quality assurance. In this study, a unique method for enhancing leather quality assurance using Convolutional Neural Networks (CNNs) is presented. Our suggested solution is able to detect and classify leather defects with amazing accuracy and efficiency by utilizing deep learning. CNN integration for leather fault categorization and detection offers a more reliable and affordable solution, ensuring better product quality and satisfied customers. Existing methods for inspecting leather products often suffer from inconsistent quality control, subjective manual inspection processes, and the complexity of accurately identifying and classifying various types of leather defects. By addressing these challenges, the proposed system aims to empower leather manufacturers with a reliable and efficient solution for ensuring consistent product quality and reducing production costs. Convolutional Neural Networks (CNNs) are integrated to provide a highly automated and effective solution that overcomes these scaling challenges. The proposed CNN-based method uses deep learning algorithms to decode images, which allows the system to swiftly and precisely analyse enormous volumes of visual data. To make sure the CNN can differentiate between various defect classes and keep a low rate of false positives and false negatives, its accuracy, precision, recall, and F1 score should be carefully assessed.

关键词： Deep learning Visualization Quality assurance Accuracy Quality control Product design Quality assessment Convolutional neural networks Reliability Defect detection

来源：评论

学校读者我要写书评

暂无评论

Frequency-Domain Auditory Suppression Modelling (FASM) - A WDFT-based anthropomorphic noise-robust feature extraction algorithm for speech recognition

Frequency-Domain Auditory Suppression Modelling (FASM) - A W...

引用

9th European Conference on Speech Communication and Technology

作者： Ivanov, Alexei V. Parfieniuk, Marek Petrovsky, Alexander A. Speech Technology Center Moscow Russia Real Time Systems Department Faculty of Computer Science Bialystok Technical University Bialystok Poland

This paper presents a physiologically inspired feature extraction algorithm for employment within the speech recognition engines, which are supposed to remain effective in noisy environments. Essentially, the algorithm simulates a key property of the "active cochlea" models - a signal dependent variable gain over the frequency range, in order to drastically reduce computational complexity of the algorithm in comparison to the original time domain "active cochlea" models, it is implemented in the frequency domain with the help of a warped discrete Fourier transformation (WDFT). The essence of FASM technique is that in the presence of the noise, higher frequency channels get more attenuation if there are "enough" signal components in the lower, less susceptible to the noise influence, part of the spectrum. As it is confirmed by the performed measurements FASM algorithm allows to boost feature invariance to noise while keeping feature informativeness at the acceptable level.

关键词： Sound recording

来源：评论

学校读者我要写书评

暂无评论

A Novel Framework of Horizontal-Vertical Hybrid Federated Learning for EdgeIoT

IEEE Networking Letters

引用

IEEE Networking Letters 2025年

作者： Li, Kai Liang, Yilei Yuan, Xin Ni, Wei Crowcroft, Jon Yuen, Chau Akan, Ozgur B. Group Department of Engineering University of Cambridge CambridgeCB3 0FA United Kingdom Real-Time and Embedded Computing Systems Research Centre CISTER Porto4249-015 Portugal The Department of Computer Science and Technology University of Cambridge CambridgeCB3 0FA United Kingdom Commonwealth Scientific and Industrial Research Organization CSIRO SydneyNSW2122 Australia The School of Computing Science and Engineering University of New South Wales Australia Nanyang Technological University SG 639798 Singapore Group Division of Electrical Engineering Department of Engineering University of Cambridge CambridgeCB3 0FA United Kingdom The Center for neXt-generation Communications CXC Koç University Istanbul34450 Turkey

This letter puts forth a new hybrid horizontal-vertical federated learning (HoVeFL) for mobile edge computing-enabled Internet of Things (EdgeIoT). In this framework, certain EdgeIoT devices train local models using the same data samples but analyze disparate data features, while the others focus on the same features using non-independent and identically distributed (non-IID) data samples. Thus, even though the data features are consistent, the data samples vary across devices. The proposed HoVeFL formulates the training of local and global models to minimize the global loss function. Performance evaluations on CIFAR-10 and SVHN datasets reveal that the testing loss of HoVeFL with 12 horizontal FL devices and six vertical FL devices is 5.5% and 25.2% higher, respectively, compared to a setup with six horizontal FL devices and 12 vertical FL devices. © 2019 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

real-time Cell Sorting with Scalable In Situ FPGA-Accelerated Deep Learning

arXiv

引用

arXiv 2025年

作者： Islam, Khayrul Forelli, Ryan F. Han, Jianzhong Bhadane, Deven Huang, Jian Agar, Joshua C. Tran, Nhan Ogrenci, Seda Liu, Yaling Department of Mechanical Engineering Lehigh University BethlehemPA18015 United States Department of Electrical and Computer Engineering Northwestern University EvanstonIL60208 United States Coriell Institute for Medical Research CamdenNJ United States Department of Computer Science Lehigh University BethlehemPA18015 United States Cooper Medical School of Rowan University CamdenNJ08103 United States Center for Metabolic Disease Research Temple University Lewis Katz School of Medicine PhiladelphiaPA19122 United States Department of Mechanical Engineering and Mechanics Drexel University PhiladelphiaPA19104 United States Real-time Processing Systems Division Fermi National Accelerator Laboratory BataviaIL60510 United States Department of Bioengineering Lehigh University BethlehemPA18015 United States

Precise cell classification is essential in biomedical diagnostics and therapeutic monitoring, particularly for identifying diverse cell types involved in various diseases. Traditional cell classification methods such as flow cytometry depend on molecular labeling which is often costly, time-intensive, and can alter cell integrity. To overcome these limitations, we present a label-free machine learning framework for cell classification, designed for real-time sorting applications using bright-field microscopy images. This approach leverages a teacher-student model architecture enhanced by knowledge distillation, achieving high efficiency and scalability across different cell types. Demonstrated through a use case of classifying lymphocyte subsets, our framework accurately classifies T4, T8, and B cell types with a dataset of 80,000 preprocessed images, accessible via an open-source Python package for easy adaptation. Our teacher model attained 98% accuracy in differentiating T4 cells from B cells and 93% accuracy in zero-shot classification between T8 and B cells. Remarkably, our student model operates with only 0.02% of the teacher model’s parameters, enabling field-programmable gate array (FPGA) deployment. Our FPGA-accelerated student model achieves an ultra-low inference latency of just 14.5 µs and a complete cell detection-to-sorting trigger time of 24.7 µs, delivering 12x and 40x improvements over the previous state-of-the-art real-time cell analysis algorithm in inference and total latency, respectively, while preserving accuracy comparable to the teacher model. This framework provides a scalable, cost-effective solution for lymphocyte classification, as well as a new SOTA real-time cell sorting implementation for rapid identification of subsets using in situ deep learning on off-the-shelf computing hardware. © 2025, CC BY.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Spatio-Temporal Anomalous Thermal Behavior Monitoring of Inside-Built Environments 19

Unsupervised Spatio-Temporal Anomalous Thermal Behavior Moni...

引用

19th Annual International Conference on Distributed Computing in Smart systems and the Internet of Things, DCOSS-IoT 2023

作者： Khan, Naima Al Hafiz Khan, Md Abdullah Roy, Nirmalya Department of Information Systems United States Umbc Center for Real-time Distributed Sensing and Autonomy United States Kennesaw State University Department of Computer Science United States

ISBN: (纸本)9798350346497

Continual wavering of outside weather degrades the efficiency of inside building envelope over time and leads to additional energy consumption, various structural damages, etc. Frequent monitoring of the indoor built environment with thermal images can assist in identifying the energy-leaking and potentially damage-prone areas. Although in recent years different researches performed deep learning and computer vision based thermal anomaly detection in built environment, several issues related to conducting strategic non-intrusive indoor thermal inspection using temporal thermal images, are still unresolved in uncontrolled environment of residential buildings. In this work, we propose a scalable thermal image-based monitoring approach for building envelopes combining the visual knowledge of structural joint information among different building components and their corresponding temporal thermal status. We collected longitudinal thermal images from indoor scenes of different building components (e.g., door, window, wall) and employed a high-level spatio-temporal graph (st-graph) to represent the structural connection among different building components and their temporal self-changes. Our proposed novel unsupervised spatio-temporal clustering framework assigns the cluster label to nodes in st-graph, combining its structural (the self and neighboring component) and temporal features which achieves better performance in identifying thermal variation compared to other clustering based approaches. We demonstrate thermal variation across the spots which indicates the potential energy leakage areas inside the built environment. The cluster patterns obtained from our proposed model assist in understanding the thermal characteristics of various surfaces at certain conditions, such as sun reflection and airflow in the inside built environment. © 2023 IEEE.

关键词： Potential energy

来源：评论

学校读者我要写书评

暂无评论

An optimization framework for dynamic, distributed real-time systems

An optimization framework for dynamic, distributed real-time...

引用

International Parallel and Distributed Processing Symposium, IPDPS 2003

作者： Ecker, Klaus Juedes, David Welch, Lonnie Chelberg, David Bruggeman, Carl Drews, Frank Fleeman, David Parrott, David Pfarr, Barbara Department of Computer Science Technical University of Clausthal Clausthal-Zellerfeld38678 Germany Center for Intelligent Distributed and Dependable Systems School of Electrical Engineering and Computer Science Ohio University AthensOH45701 United States Real-Time Software Engineering Branch NASA Goddard Space Flight Center BaltimoreMD20771 United States

ISBN: (纸本)0769519261

The paper presents a model that is useful for developing resource allocation algorithms for distributed real-time systems that operate in dynamic environments. Interesting aspects of the model include dynamic environments, utility and service levels, which provide a means for graceful degradation in resource-constrained situations and support optimization of the allocation of resources. The paper also provides an allocation algorithm that illustrates how to use the model for producing feasible, optimal resource allocations. © 2003 IEEE.

关键词： Constrained optimization

来源：评论

学校读者我要写书评

暂无评论

Analysing multimedia traffic in real-time ATM networks

Analysing multimedia traffic in real-time ATM networks

引用

real-time Technology and Applications Symposium

作者： M. Sjodin H. Hansson Department of Computer Systems University of Uppsala Sweden Mälardalen Real-Time Research Center Mälardalen University Sweden

We present a method for providing response-time guarantees for multimedia traffic in Asynchronous Transfer Mode (ATM) networks. The method extends our Response-time Analysis (RTA) for ATM networks to better capture the behaviour of MPEG-type multimedia traffic. RTA is intended for admission control of hard real time traffic, i.e., it determines if a new connection can be admitted without violating the timing requirements of the new as well as any already admitted connection. We evaluate the extended RTA by comparing it with our original RTA, which does not consider the specific traffic patterns of MPEG-type multimedia traffic. In addition, we show how Weighted Fair Queueing (WFQ) and Calculus for Network Delays (CND) can be adopted to MPEG-type multimedia traffic, and compare the thus obtained analysis with our extended RTA. Evaluations indicate that our new method yields both higher achievable network utilisation and shorter analysis times compared to the original RTA, and higher achievable network utilisation compared to both the adapted WFQ and CND.

关键词： Telecommunication traffic Intelligent networks Asynchronous transfer mode Traffic control ISDN Timing Pattern analysis real time systems Tellurium Content addressable storage

来源：评论

学校读者我要写书评

暂无评论

Schedulability analysis of multi-packet messages in segmented CAN

Schedulability analysis of multi-packet messages in segmente...

引用

2012 IEEE 17th International Conference on Emerging Technologies and Factory Automation, ETFA 2012

作者： Azketa, Ekain Gutierrez, J. Javier Palencia, J. Carlos Harbour, Michael Gonzalez Almeida, Luis Marcos, Marga Ikerlan Technology Research Center Software Technologies Department Mondragón Spain University of Cantabria Computers and Real-Time Group Santander Spain University of Porto Electrical and Computer Engineering Department Porto Portugal University of Basque Country Systems Engineering and Automation Department Bilbao Spain

ISBN: (纸本)9781467347372

The CAN bus is one of the most used networks in distributed real-time systems. The CAN bus can be divided by means of bridges in order to distribute the network load among different segments and contribute positively to the schedulability of the system. On the other hand, CAN supports the fragmentation of a message in multiple packets when it does not fit into one single CAN frame. The schedulability analysis of CAN has been extensively studied in the literature. Although the proposed methods analyze non-segmented CAN buses, they can be applied to segmented networks. However, the techniques from the related literature obtain quite pessimistic results in the analysis of segmented CAN networks with multi-packet messages. This paper proposes a less pessimistic CAN schedulability analysis method for the mentioned kind of networks. Moreover, this optimized technique can be integrated with methods oriented to the holistic schedulability analysis of a complete distributed real-time system. © 2012 IEEE.

关键词： Interactive computer systems

来源：评论

学校读者我要写书评

暂无评论

Generating robot action plans by means of an heuristic search

Generating robot action plans by means of an heuristic searc...

引用

IEEE International Conference on Robotics and Automation (ICRA)

作者： B. Frommherz G. Werling Campus-based Engineering Center (CEC) Digital Equipment GmbH Karlsruhe Germany Institute of Real-time Computer Systems and Robotics University of Karlsruhe Germany

A planning concept is proposed which meets the requirements of flexibility and simplicity of the resulting plan structures and efficiency of replanning. The concept is based on a heuristic search process (branch-and-bound) by which a plan generation tree containing (as nodes) the most promising solution graphs is partly developed. This is achieved by a heuristic evaluation function which estimates the quality of the intermediate plans during the plan synthesis phase. The concept is shown for the area of automatic assembly using a small assembly task as an example. A short description of a robot planning system which is based on the concept is included.< >

关键词： Robots Robotic assembly Power system planning real time systems Meeting planning Phase estimation Robotics and automation Assembly systems Artificial intelligence Search methods

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：