检索结果-内蒙古大学图书馆

A gradient algorithm for stereo matching without correspondence

IEEE TRANSACTIONS ON AUTOMATIC CONTROL 1996年第11期41卷 1671-1676页

作者： Zhou, J Ghosh, BK WASHINGTON UNIV DEPT SYST SCI & MATHST LOUISMO 63130

In this paper we study the problem of feature point matching via a technique well known in geometry, leading to a new gradient algorithm in machine vision. The procedure does not require one to solve explicit correspondences between feature points but relies on a specific form of ''spatial averaging.'' In this sense the estimation procedure is robust. In this paper we describe the algorithm and report simulation results.

关键词： Lie algebras computer vision feature extraction image matching optimisation stereo image processing Lie algebra feature point matching gradient algorithm machine vision spatial averaging stereo correspondence stereo matching Lie algebras Feature Points computer vision image matching algorithms assessment procedure Stereo image processing robot vision Correspondence Feature extraction

来源：评论

学校读者我要写书评

暂无评论

gradient algorithm with perturbations for localization in WSNs 35

Gradient algorithm with perturbations for localization in WS...

引用

第35届中国控制会议

作者： LIU Hui TIAN Yu-Ping School of Automation Southeast University

ISBN: (纸本)9781509009107

This paper investigates how to determine the nodes' locations in wireless sensor networks(WSNs) with only distance *** extend our formation algorithm given in[1]to this localization *** extension includes three terms:the estimation of initial position,time-varying step size and decaying *** is verified that excellent performance can be derived,and the global convergence is maintained under this new localization *** this protocol uses the neighbors' information without any global knowledge,it is a fully distributed *** simulations are provided to illustrate the effectiveness of the proposed method.

关键词： wireless sensor network gradient algorithm variable step size

来源：评论

学校读者我要写书评

暂无评论

gradient algorithm for structural parameter estimation and nonlinear restoring forces

Gradient algorithm for structural parameter estimation and n...

引用

American Control Conference

作者： Garrido, R Rivero-Angeles, FJ Martinez-Garcia, JC Martinez-Guerra, R Gomez-Gonzalez, B IPN CINVESTAV Dept Automat Control Mexico City 07300 DF Mexico

ISBN: (纸本)0780383354

Civil structures undergo hysteresis cycles due to cracking or yielding when subjected to severe strong motions. Seismic instrumentation, structural monitoring and system identification techniques have been used in the past years to assess damage. The present research proposes a gradient algorithm to simultaneously identify the combined driving and nonlinear restoring forces, and structural parameters, such as damping and stiffness in a single-degree-of-freedom structure. Simulations are carried out using the El Centro (1940) and the Mexico City (1985) seismic records. Encouraging results are obtained for the identification of the parameters and the estimation of the driving and restoring forces.

关键词： gradient methods structural engineering parameter estimation condition monitoring computerised monitoring nonlinear restoring force structural parameter estimation gradient algorithm civil structure hysteresis cycle seismic instrumentation structural monitoring system identification technique driving force

来源：评论

学校读者我要写书评

暂无评论

gradient algorithm Based ISAR Image Reconstruction From the Incomplete Dataset 3

Gradient Algorithm Based ISAR Image Reconstruction From the ...

引用

3rd International Workshop on Compressed Sensing Theory and its Application to Radar, Sonar, and Remote Sensing (CoSeRa)

作者： Dakovic, Milos Stankovic, Ljubisa Stankovic, Srdjan Univ Montenegro Fac Elect Engn Podgorica 81000 Montenegro

ISBN: (纸本)9781479974207

In the inverse synthetic aperture radar (ISAR) number of target reflectors is small resulting in the fact that ISAR images are sparse. Since the ISAR image is obtained using the Fourier transform of the input signal then this problems can be treated within processing of sparse signals. The compressive sensing (CS) theory proves that, under some conditions, exact reconstruction of sparse signals is possible based on the reduced set of observations. Here we will present a gradient based reconstruction algorithm and apply it to the several ISAR setups. In contrast to the common reconstruction algorithms where the signal in its sparsity domain is reconstructed, this algorithm solves minimization in an indirect way, by calculating the missing samples/measurements. Obtained results show that the presented simple reconstruction algorithm is reliable and robust. Common problems in ISAR imaging, like uncompensated motion or target nonuniform motion, make ISAR image only approximately sparse. The presented procedure can provide useful results in these cases as well.

关键词： ISAR Compressive sensing gradient algorithm

来源：评论

学校读者我要写书评

暂无评论

Subpixel displacement measurement method based on the combination of particle swarm optimization and gradient algorithm

引用

OPTICAL ENGINEERING 2017年第10期56卷 104101-104101页

作者： Chen Guang Feng Qibo Ding Keqin Gao Zhan Beijing Jiaotong Univ Beijing Peoples R China China Special Equipment Inspect & Res Inst Beijing Peoples R China

A subpixel displacement measurement method based on the combination of particle swarm optimization (PSO) and gradient algorithm (GA) was proposed for accuracy and speed optimization in GA, which is a subpixel displacement measurement method better applied in engineering practice. An initial integer-pixel value was obtained according to the global searching ability of PSO, and then gradient operators were adopted for a subpixel displacement search. A comparison was made between this method and GA by simulated speckle images and rigid-body displacement in metal specimens. The results showed that the computational accuracy of the combination of PSO and GA method reached 0.1 pixel in the simulated speckle images, or even 0.01 pixels in the metal specimen. Also, computational efficiency and the antinoise performance of the improved method were markedly enhanced. (C) 2017 Society of Photo-Optical Instrumentation Engineers (SPIE)

关键词： digital image correlation subpixel search particle swarm optimization gradient algorithm combination of particle swarm optimization and gradient algorithm

来源：评论

学校读者我要写书评

暂无评论

Steady-state analysis of a complex adaptive notch filter using modified gradient algorithm

引用

AEU-INTERNATIONAL JOURNAL OF ELECTRONICS AND COMMUNICATIONS 2014年第11期68卷 1112-1118页

作者： Punchalard, R. Mahanakorn Univ Technol Dept Telecommun Engn Bangkok 10530 Thailand

Based on power spectral density (PSD) analytical technique, mean square error (MSE) (or variance) of the frequency estimate of a first-order complex adaptive IIR notch filter (ANF) using modified complex plain gradient (MCPG) algorithm is investigated in this paper. The steady-state expression for MSE is derived in closed form. A quantitative analysis for the estimation MSE has been carried out. It has been revealed that the MSE of frequency estimate is independent of an input frequency of a complex sinusoid. In addition, computer simulations are treated to corroborate the theoretical analysis and the relationships between MSE and system parameters are shown. (C) 2014 Elsevier GmbH. All rights reserved.

关键词： Complex notch filter gradient algorithm Frequency estimation Power spectral density

来源：评论

学校读者我要写书评

暂无评论

Switching neuronal state: optimal stimuli revealed using a stochastically-seeded gradient algorithm

引用

JOURNAL OF COMPUTATIONAL NEUROSCIENCE 2014年第3期37卷 569-582页

作者： Chang, Joshua Paydarfar, David Univ Massachusetts Sch Med Dept Neurol Worcester MA 01610 USA Harvard Univ Wyss Inst Biologically Inspired Engn Boston MA 02115 USA

Inducing a switch in neuronal state using energy optimal stimuli is relevant to a variety of problems in neuroscience. Analytical techniques from optimal control theory can identify such stimuli;however, solutions to the optimization problem using indirect variational approaches can be elusive in models that describe neuronal behavior. Here we develop and apply a direct gradient-based optimization algorithm to find stimulus waveforms that elicit a change in neuronal state while minimizing energy usage. We analyze standard models of neuronal behavior, the Hodgkin-Huxley and FitzHugh-Nagumo models, to show that the gradient-based algorithm: 1) enables automated exploration of a wide solution space, using stochastically generated initial waveforms that converge to multiple locally optimal solutions;and 2) finds optimal stimulus waveforms that achieve a physiological outcome condition, without a priori knowledge of the optimal terminal condition of all state variables. Analysis of biological systems using stochastically-seeded gradient methods can reveal salient dynamical mechanisms underlying the optimal control of system behavior. The gradient algorithm may also have practical applications in future work, for example, finding energy optimal waveforms for therapeutic neural stimulation that minimizes power usage and diminishes off-target effects and damage to neighboring tissue.

关键词： Computational model gradient algorithm Optimal stimulation Single-cell neuron

来源：评论

学校读者我要写书评

暂无评论

A multi-innovation generalized extended stochastic gradient algorithm for output nonlinear autoregressive moving average systems

引用

APPLIED MATHEMATICS AND COMPUTATION 2014年 247卷 218-224页

作者： Hu, Yuanbiao Liu, Baolin Zhou, Qin China Univ Geosci Minist Land & Resources Key Lab Deep GeoDrilling Technol Beijing 100083 Peoples R China

This paper proposes a generalized extended stochastic gradient (GESG) algorithm for estimating the parameters of a class of Wiener nonlinear autoregressive moving average systems using the gradient search. In order to improve the convergence rates of the GESG algorithm, a multi-innovation GESG algorithm is derived. The simulation results indicate that the proposed algorithms can effectively estimate the parameters of a class of output nonlinear systems. (C) 2014 Elsevier Inc. All rights reserved.

关键词： Parameter estimation Recursive identification gradient algorithm Nonlinear system

来源：评论

学校读者我要写书评

暂无评论

A new memoryless nonlinear gradient algorithm for a second-order adaptive IIR notch filter and its performance analysis

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-ANALOG AND DIGITAL SIGNAL PROCESSING 1998年第4期45卷 462-472页

作者： Xiao, YG Tadokoro, Y Kobayashi, Y Saga Univ Fac Sci & Engn Dept Elect & Elect Engn Saga 840 Japan Toyohashi Univ Technol Fac Engn Dept Informat & Comp Sci Toyohashi Aichi 441 Japan Denso Create Inc Nagoya Aichi 460 Japan

Adaptive infinite inpulse-response (IIR) notch filters have been widely studied for many years. However, further efforts are still to be made to pursue new algorithms which work better than the conventional plain gradient algorithm but have little increase in complexity. In this paper, we employ the gradient linearization, Taylor series expansion, and calculus of variations to derive a novel memoryless nonlinear gradient algorithm for a second-order adaptive IIR notch filter, which improves the estimation performance considerably. Approximate closed-form expressions for the stability bounds on the step size parameter and the steady-state coefficient variance are also derived. Extensive simulations indicate the significant improvement that may be achieved using the new algorithm, and verify the validity of the analytical results.

关键词： frequency estimation gradient algorithm infinite impulse-response notch filtering performance analysis

来源：评论

学校读者我要写书评

暂无评论

Convergence of a Batch gradient algorithm with Adaptive Momentum for Neural Networks

引用

NEURAL PROCESSING LETTERS 2011年第3期34卷 221-228页

作者： Shao, Hongmei Xu, Dongpo Zheng, Gaofeng China Univ Petr Sch Math & Comput Sci Dongying 257061 Peoples R China Harbin Engn Univ Coll Sci Harbin 150001 Peoples R China Rakuten Inc Platform Search Grp Shinagawa Tokyo 1400002 Japan

In this paper, a batch gradient algorithm with adaptive momentum is considered and a convergence theorem is presented when it is used for two-layer feedforward neural networks training. Simple but necessary sufficient conditions are offered to guarantee both weak and strong convergence. Compared with existing general requirements, we do not restrict the error function to be quadratic or uniformly convex. A numerical example is supplied to illustrate the performance of the algorithm and support our theoretical finding.

关键词： Neural network gradient algorithm Adaptive momentum Convergence

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：