检索结果-内蒙古大学图书馆

Asian conference on Pattern Recognition (ACPR)

作者： Chuan-Xian Ren Dao-Qing Dai Hong Yan Center of Computer Vision and Department of Mathematics Sun Yat-Sen University Guangzhou China Department of Electric Engineering City University of Hong Kong Hong Kong China

Practical face recognition systems are sometimes confronted with low-resolution (LR) images. Most existing feature extraction algorithms aim to preserve relational structure among objects of the input space in a linear embedding space. However, it has been a consensus that such complex visual learning tasks will be well be solved by adopting multiple descriptors to more precisely characterize the data for improving performance. In this paper, we addresses the problem of matching LR and high-resolution images that are difficult for conventional methods in practice due to the lack of an efficient similarity measure, and a multiple kernel criterion (MKC) is proposed for LR face recognition without any super-resolution (SR) preprocessing. Different image descriptors including RsL2, LBP, Gradientface and IMED are considered as the multiple kernel generators and the Gaussian function is exploited as the distance induced kernel. MKC solves this problem by minimizing the inconsistency between the similarities captured by the multiple kernels, and the nonlinear objective function can be alternatively minimized by a constrained eigenvalue decomposition. Experiments on benchmark databases show that our MKC method indeed improves the recognition performance.

关键词： Kernel Image resolution Face recognition Face Feature extraction Databases optimization

来源：评论

学校读者我要写书评

暂无评论

Application of SAO-Optimized Hybrid Neural Network Model in Stock Price Prediction

Application of SAO-Optimized Hybrid Neural Network Model in ...

引用

International Symposium on Environmentally Conscious Design and Inverse Manufacturing (EcoDesign)

作者： Mingwei Liang Xiaoxiao Li School of science Shanghai Institute of Technology College of Sciences Shanghai China

ISBN: (数字)9798350363043

ISBN: (纸本)9798350363050

With the increasing complexity of global financial markets, traditional stock price prediction methods often perform poorly due to their inability to effectively handle nonlinear and high-dimensional data. Therefore, this paper constructs a hybrid SAO-CNN-BiLSTM-SA model, which uses convolutional neural networks (CNN) to automatically extract features from time series data, bi-directional long and short-term memory networks (BiLSTM) to capture temporal dependencies, self-attention self-attention mechanism (SA) to enhance understanding of global dependencies, and the SAO algorithm to optimize hyper parameters. The model is trained and tested using historical minute-level data from the Shanghai Stock Exchange Composite Index and compared with traditional CNN-BiLSTMSA models. Experimental results show that the SAO-CNN-BiLSTM-SA model significantly outperforms traditional models in key performance indicators such as root mean square error (RMSE), mean absolute error (MAE), and R-squared value. Especially in handling sudden market events and high volatility, the new model shows greater adaptability and accuracy. The successfully developed SAO-CNN-BiLSTM-SA model in this study not only improves stock price prediction accuracy but also demonstrates its application potential in complex financial data analysis by integrating the latest machine learning techniques and optimization algorithms.

关键词： Adaptation models Analytical models Accuracy Time series analysis Predictive models Feature extraction Prediction algorithms Data models Convolutional neural networks optimization

来源：评论

学校读者我要写书评

暂无评论

Research on Robot Path Planning and Localization Based on ROS

Research on Robot Path Planning and Localization Based on RO...

引用

Computer, Big Data and Artificial Intelligence (ICCBD+AI), International conference on

作者： Wenhan Ji Shuai Wu Minze Wang Chengge Duan JunYang Yang Jun Lu Peng Cheng Kainan Xu Suzhou Hanyun Keling Network Information Service Co. Ltd. Suzhou China China Mobile Communications Group Jiangsu Co. Ltd. Suzhou Branch Suzhou China School of Advanced Technology Xi’an Jiaotong - Liverpool University Suzhou China Southeast University Nanjing China Relight Studio iQIYI Inc. Shanghai China Suzhou PinYi Film and Television Co. Ltd. Suzhou China HENAN University School of Software Kaifeng China

ISBN: (数字)9798331533991

ISBN: (纸本)9798331534004

Recently, the rapid development of robotics technology has made enhancing autonomous navigation capabilities in complex environments increasingly essential. Path planning is key to improving the autonomy of robots, while localization is the process of determining the robot’s accurate position within the environment. Given this, this research delves into path planning and localization technologies based on the Robot Operating System (ROS), comparing and analyzing path planning algorithms such as Dijkstra, A*, RRT and its variants, as well as particle filter localization techniques. By discussing the performance of these technologies in practical applications, we highlight their strengths and weaknesses in high-dimensional spaces and complex terrains and proposes corresponding optimization strategies to enhance robots’ navigation capabilities in complex environments. The research indicates that the appropriate combination of various path planning and localization technologies in different application scenarios can effectively improve the performance of robotic systems, providing a theoretical basis and practical guidance for the optimization of robotic navigation systems.

关键词： Location awareness Smoothing methods Operating systems Particle filters Robot localization Trajectory Planning Robots optimization Autonomous robots

来源：评论

学校读者我要写书评

暂无评论

A high-performance RISC-V SoC with a DNN Accelerator for Energy Applications

A High-Performance RISC-V SoC with a DNN Accelerator for Ene...

引用

Electronic Information Engineering and Computer Communication (EIECC), International conference on

作者： Yang Ren Xinyu Liu Lewei Li Nianxiong Tan College of Integrated Circuits Zhejiang University Hangzhou China Research and Development Center Hangzhou Vango Technologies Inc Hangzhou China College of Integrated Circuits Zhejiang University Hangzhou Vango Technologies Inc Hangzhou China

ISBN: (数字)9798331534622

ISBN: (纸本)9798331534639

As in many applications, the energy sector is witnessing the migration of artificial intelligence (AI) algorithms from the cloud to the edge. This paper introduces a System-on-Chip (SoC) designed for energy control units (ECUs) and data concentrators (DCs) to support edge computing acceleration. The SoC is implemented on the Chipyard platform, utilizing the open-source out-of-order processor BOOM and the software-friendly accelerator Gemmini. Fabricated using a 22nm process, the core achieves a normal operating frequency of 1GHz and occupies a die area of 22.99mm 2 .

关键词： Out of order AI accelerators System-on-chip Smart grids optimization Next generation networking Edge computing

来源：评论

学校读者我要写书评

暂无评论

optimization of speed control algorithm to achieve minimum torque ripple for a switched reluctance motor drive via GA

Optimization of speed control algorithm to achieve minimum t...

引用

International conference on Power Electronics Systems and Applications (ICPESA)

作者： Hanif Tahersima Mohammadjafar Kazemsaleh Mohammadhossein Tahersima Navid Hamedi Process Engineering Development Research Institute of Petroleum Industry (RIPI) Tehran Iran Electrical Engineering Department Iran University of Science and Technology Tehran Iran Electronic System Engineering Kyoto Institute of Technology Kyoto Japan

This study presents a speed controller design for a switched reluctance (SR) motor in order to achieve minimum torque ripple and high control performance. First of all, SR motor convertor designed for soft chopping is chosen. This converter as well as producing less torque ripple, provides more degrees of freedom for SR motor drive controller. A PID controller and a switching algorithm for turn-on and turn-off degree of each phase of motor form speed control loop of SR motor drive. The primary parameters of controller are achieved by try and error. But eventually an optimization algorithm to reach the goals and constraints in different set points is defined and its parameters are optimized with a Genetic Algorithm (GA). This algorithm optimized the turn-on and turn-off degrees of each phase, the parameters of PID controller in transient state, and parameters of PID controller that considered for reducing the torque ripple in steady state. Also, GA simultaneously obtains the optimum parameters of three nonlinear gains considered for fuzzy switching between the two PID controllers. The proposed control algorithm was simulated using MATLAB / Simulink software package and an application example of 8/6 SRM to validate the performance of designed algorithm.

关键词： Reluctance motors Torque Strontium Load modeling Switches Steady-state Transient analysis

来源：评论

学校读者我要写书评

暂无评论

FDRC: Flow-Driven Rule Caching optimization in software Defined Networking

FDRC: Flow-Driven Rule Caching Optimization in Software Defi...

引用

IEEE International conference on Communications

作者： He Li Song Guo Chentao Wu Jie Li School of Computer Science and Engineering The University of Aizu Shanghai Key Laboratory of Scalable Computing and Systems Department of Computer Science & Engineering Shanghai Jiao Tong University

ISBN: (纸本)9781467364300

With the sharp growth of cloud services and their possible combinations, the scale of data center network traffic has an inevitable explosive increasing in recent years. software defined network (SDN) provides a scalable and flexible structure to simplify network traffic management. It has been shown that Ternary Content Addressable Memory (TCAM) management plays an important role on the performance of SDN. However, previous literatures, in point of view on rule placement strategies, are still insufficient to provide high scalability for processing large flow sets with a limited TCAM size. So caching is a brand new method for TCAM management which can provide better performance than rule placement. In this paper, we propose FDRC, an efficient flow-driven rule caching algorithm to optimize the cache replacement in SDN-based networks. Different from the previous packet-driven caching algorithm, FDRC is characterized by trying to deal with the challenges of limited cache size constraint and unpredictable flows. In particular, we design a caching algorithm with low-complexity to achieve high cache hit ratio by prefetching and special replacement strategy for predictable and unpredictable flows, respectively. By conducting extensive simulations, we demonstrate that our proposed caching algorithm significantly outperforms FIFO and least recently used (LRU) algorithms under various network settings.

关键词： Flexible structures Prefetching Switches Silicon

来源：评论

学校读者我要写书评

暂无评论

Wind Direction Prediction Based on nonlinear Autoregression and Elman Neural Networks for the Wind Turbine Yaw System

SSRN

引用

SSRN 2022年

作者： Yang, Yusong Solomin, Evgeny V. Shishkov, Aleksandr N. Department of Electric Stations Grids and Systems of Power Supply South Ural State University 76 Lenin prospekt Chelyabinsk454080 Russia Department of Electrical Engineering Moscow Polytechnic University Moscow Russia

At present, about 90% of wind power is represented by big megawatt class horizontal axis wind turbines (HAWT). One of the permanent modern trends is the optimization and improvement of the HAWT yaw control algorithms. However, still none of the proposed yaw control strategies can give a hundred percent result of power output. Thus, the research and development in this area *** this article we proposed a model for predicting a wind direction based on nonlinear autoregression (NAR) and Elman neural networks (ENN), which is designed to reduce the error in determination of wind *** model has demonstrated a high prediction accuracy of wind direction determination (up to 1.79°), comparing with the results obtained by traditional instruments (exceeding 5°). The modeling in Matlab/Simulink software package confirmed that the neural network based strategy can accurately and quickly predict changes in wind direction, which significantly improves the accuracy and performance of the HAWT yaw system. This not only reduces the yawing process time to a certain extent, but also allows to effectively reduce the number of yawing in time and extend the lifetime of wind turbine mechanical components, at the same time increasing the output power on more than 6%. © 2022, The Authors. All rights reserved.

关键词： Wind power

来源：评论

学校读者我要写书评

暂无评论

Research on the Design of LLC Digital Power Supply Based on Microcontroller

Research on the Design of LLC Digital Power Supply Based on ...

引用

International Symposium on Environmentally Conscious Design and Inverse Manufacturing (EcoDesign)

作者： Chunguang Zhou Jian Liu School of Electrical and Electronic Engineering Shanghai Institute of Technology Shanghai China

ISBN: (数字)9798350363043

ISBN: (纸本)9798350363050

In this study, we have undertaken a comprehensive design of a digital power supply, leveraging the LLC resonant converter and controlled by a cost-effective STMicroelectronics’s microcontroller. Recognizing the limitations of traditional analog power systems in delivering high-efficiency, compact, and intelligent solutions, we have embraced digital control technology to address the challenges of output stability inherent in the LLC resonant converter. Our approach offers precise control, rapid response, and a reduced environmental footprint, significantly enhancing the stability and reliability of the power supply system. We have detailed the design and implementation of our LLC digital power supply. Our design includes self-tuning, cost-effectiveness, improved dynamic response, and self-monitoring capabilities through custom algorithms. We have calculated and selected the models of key components, such as optocouplers and transformers, in the inverter circuit for optimal performance. The software system is based on STMicroelectronics’s embedded microcontroller and is divided into PWM control, ADC reading, PID closed-loop control algorithm and error handling to achieve stable voltage output. The successful development of a 1000 W LLC digital power supply paves the way for future advancements in control strategies, component optimization, and environmental adaptability in power supply systems.

关键词： Power system measurements Power supplies Microcontrollers Density measurement Heuristic algorithms software algorithms Resonant converters Power system stability Stability analysis Circuit stability

来源：评论

学校读者我要写书评

暂无评论

A Secure Aware Routing Protocol for Efficient and Reliable FPANETS

A Secure Aware Routing Protocol for Efficient and Reliable F...

引用

software, Systems and Information Technology (SSITCON), International conference on

作者： P Arockia Mary D Maalini K Manivannan K Umamaheswari S. Balaji R Deepak Department of Information Technology V.S.B. Engineering College Karur India

ISBN: (数字)9798350352931

ISBN: (纸本)9798350352948

The use of Unmanned Aerial Vehicles (UAVs) has evolved significantly, with a current focus on utilizing multiple UAVs in teams to achieve high-level objectives. To support this evolution, new networking frameworks are needed for the highly mobile nodes like UAVs in fleets. These frameworks enable both indirect and direct communication between nodes within communication range. However, constructing an ad hoc network between UAVs presents challenges, with criteria that differ from traditional networks such as Mobile Ad-hoc Networks (MANETs) and aerial ad-hoc networks using the Flower Pollination Algorithm Protocol (FPAP). This study presents network models of UAVs, highlights open research questions, explores opportunities and future directions, and discusses the challenges of deploying UAVs as ad hoc relay nodes. The research aims to enhance data transmission security in FPANETs to reduce the risks of black hole attacks. An Artificial Neural Network (ANN) model achieves 98.97% detection accuracy, while cluster-based routing streamlines network architecture and reduces complexity. performance evaluations consider end-to-end latency, network throughput, hop count, and other network metrics. This paper highlights the effectiveness of these techniques in FPANET security and routing optimization.

关键词： software algorithms Flower pollination algorithm Artificial neural networks Routing Throughput software systems Routing protocols Ad hoc networks Security Vehicle dynamics

来源：评论

学校读者我要写书评

暂无评论

Semi-Supervised Random Forests

Semi-Supervised Random Forests

引用

International conference on Computer Vision (ICCV)

作者： Christian Leistner Amir Saffari Jakob Santner Horst Bischof Institute for Computer Graphics and Vision Graz University of Technology Austria

Random Forests (RFs) have become commonplace in many computer vision applications. Their popularity is mainly driven by their high computational efficiency during both training and evaluation while still being able to achieve state-of-the-art accuracy. This work extends the usage of Random Forests to Semi-Supervised Learning (SSL) problems. We show that traditional decision trees are optimizing multi-class margin maximizing loss functions. From this intuition, we develop a novel multi-class margin definition for the unlabeled data, and an iterative deterministic annealing-style training algorithm maximizing both the multi-class margin of labeled and unlabeled samples. In particular, this allows us to use the predicted labels of the unlabeled data as additional optimization variables. Furthermore, we propose a control mechanism based on the out-of-bag error, which prevents the algorithm from degradation if the unlabeled data is not useful for the task. Our experiments demonstrate state-of-the-art semi-supervised learning performance in typical machine learning problems and constant improvement using unlabeled data for the Caltech-101 object categorization task.

关键词： Semisupervised learning Iterative algorithms Computer vision Application software Computational efficiency Decision trees Annealing Error correction Machine learning algorithms Degradation

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：