检索结果-内蒙古大学图书馆

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Kun Gai Zhenwei Shi Changshui Zhang State Key Laboratory on Intelligent Technology and Systems Tsinghua National Laboratory for Information Science and Technology (TNList) Department of Automation Tsinghua University Beijing China Image Processing Center School of Astronautics Beihang University Beijing China

We consider the blind separation of source layers from superimposed mixtures thereof, involving unknown motions and unknown mixing coefficients of layers in each mixture. Previous blind separation approaches for such problems assume motions to be uniform translations, and hence are limited for real world applications. In this paper, we develop a sparse blind separation algorithm to estimate both parameterized motions and mixing coefficients. Then, a novel reconstruction approach is presented to recover all layers, by utilizing not only the mixing model but also the statistical properties of natural images. The whole method can handle more general motions than translations, including scalings, rotations and other transformations. In addition, the number of layers is automatically identified, and all layers can be recovered even in the under-determined case where mixtures are fewer than layers. The effectiveness of this technology is shown in the experiments on two simulated mixtures of four layers, real photos containing transparency and reflections, and real crossfade images from videos.

关键词： Image reconstruction Motion estimation Optical reflection Cameras Laboratories Space technology Parameter estimation Videos Glass Layout

来源：评论

学校读者我要写书评

暂无评论

A fast mean-field method for large-scale high-dimensional data and its application in colonic polyp detection at CT colonography

A fast mean-field method for large-scale high-dimensional da...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Shijun Wang Ronald M. Summers Changshui Zhang Department of Radiology and Imaging Sciences National Institutes of Health DHHS Bethesda MD USA State Key Laboratory of Intelligent Technologies and Systems Department of Automation Tsinghua National Laboratory for Information Science and Technology Tsinghua University Beijing China

In this paper, we propose a fast mean-field method called LHMF to handle probabilistic models of large-scale data in high dimensional space. By using diffusion map locally linear embedding method which is a non-linear dimensionality reduction method, we first embed the high dimensional data into a low dimensional space. Then we construct a coarse-grained graph which preserves the spectral properties of original weighted graph in the high dimensional space by clustering. A new spin model is defined in the diffusion space and the geometric centroids of clusters represent variables in the new spin model. The calculation demand of mean-field methods can be reduced greatly on the coarse-grained spin model. The final marginal moments of original variables are derived from the states of geometric centroids by using geometric harmonics. We first tested the proposed method on the MNIST hand-written digits dataset. Experimental results show that the LHMF method is competent with consistency approach, a state-of-the-art semi-supervised learning method. Then we applied the proposed method to a large-scale colonic polyp dataset from computed tomography (CT) scans. Free-response operator characteristic analysis shows that our method achieves higher sensitivity with lower false positive rate compared with support vector machines.

关键词： Large-scale systems Colonic polyps Virtual colonoscopy Solid modeling Computed tomography Semisupervised learning Physics Microscopy Laboratories Neural networks

来源：评论

学校读者我要写书评

暂无评论

Motion planning with obstacle avoidance for kinematically redundant manipulators based on two recurrent neural networks

Motion planning with obstacle avoidance for kinematically re...

引用

IEEE International Conference on systems, Man and Cybernetics

作者： Xiaolin Hu Jun Wang Bo Zhang Department of Computer Science and Technology Tsinghua University Beijing China State Key Laboratory of Intelligent Technology and Systems Tsinghua National Laboratory for Information Science and Technology Beijing China Department of Mechanical and Automation Engineering Chinese University of Hong Kong New Territories Hong Kong China

Inverse kinematic motion planning of redundant manipulators by using recurrent neural networks in the presence of obstacles and uncertainties is a real-time nonlinear optimization problem. To tackle this problem, two subproblems should be resolved in real time. One is the determination of critical points on a given manipulator closest to obstacles, and the other is the computation of joint velocities of the manipulator which can direct the manipulator following a desired trajectory and away from obstacles if it is getting close to them. Different from our previous approaches where the critical points on the manipulator were assumed to be known, these points are to be computed by using a recurrent neural network in the paper. A time-varying quadratic programming problem is formulated for avoiding polyhedral obstacles. In view that the problem is not strictly convex, an existing recurrent neural network, general projection neural network, is applied for solving it. By introducing a velocity smoothing technique into our previous quadratic programming formulation of the joint velocity assignment problem, a recently developed recurrent neural network, improved dual neural network, is proposed to solve it, which features lower structural complexity compared with existing neural networks. Moreover, The effectiveness of the proposed neural networks is demonstrated by simulations on the Mitsubishi PA10-7C manipulator.

关键词： Recurrent neural networks Neural networks Quadratic programming Manipulators Laboratories Kinematics Smoothing methods Paper technology Information science technology planning

来源：评论

学校读者我要写书评

暂无评论

一种改进的利用频移估计Q值的方法

一种改进的利用频移估计Q值的方法

引用

CPS/SEG北京2009国际地球物理会议

作者： Ning Tu Wen-kai Lu State Key Laboratory of Intelligent Technology and Systems Tsinghua National Laboratory for Information Science and TechnologyDepartment of AutomationTsinghua University

现有的Q 值估计方法中，主要有两种利用振幅谱频移的方法：质心频率频移法和峰值频率频移法。这两种方法各有优缺点。在本篇文章中，我们详细的讨论并比较这两种方法，并在此基础上提出一种新的利用频移进行Q 值估计的方法。这种方法综... 详细信息

现有的Q 值估计方法中，主要有两种利用振幅谱频移的方法：质心频率频移法和峰值频率频移法。这两种方法各有优缺点。在本篇文章中，我们详细的讨论并比较这两种方法，并在此基础上提出一种新的利用频移进行Q 值估计的方法。这种方法综合了现有两种方法的优点。合成数据实验证实了这种方法的有效性。

关键词： Q值估计质心频率峰值频率频移法

来源：评论

学校读者我要写书评

暂无评论

基于非高斯性最大化的鬼波压制

基于非高斯性最大化的鬼波压制

引用

CPS/SEG北京2009国际地球物理会议

作者： Xuan Mo Wen-kai Lu State Key Laboratory of Intelligent Technology and Systems Tsinghua National Laboratory for Information Science and TechnologyDepartment of AutomationTsinghua University

在海洋地震信号处理中，鬼波是一种常见的不利的效应。本文提出了一种基于非高斯性最大化的参数搜索的方法来估计海面反射系数和鬼波时延这两个参数。通过这两个参数，我们设计鬼波滤波器来对接收到的地震信号进行反卷积，从而得到拥有... 详细信息

在海洋地震信号处理中，鬼波是一种常见的不利的效应。本文提出了一种基于非高斯性最大化的参数搜索的方法来估计海面反射系数和鬼波时延这两个参数。通过这两个参数，我们设计鬼波滤波器来对接收到的地震信号进行反卷积，从而得到拥有更多信息、更高分辨率的地震信号。文中，该方法对人工合成数据和实际数据都有很好的效果。

关键词：鬼波压制非高斯性最大化海洋地震信号处理海面反射系数

来源：评论

学校读者我要写书评

暂无评论

Routing Protocol for Wireless Sensor Networks in Home automation

引用

IFAC Proceedings Volumes 2009年第3期42卷 27-33页

作者： Xiao Hui Li Hong Qin Xu Seung Ho Hong Zhi Wang Xiang Fan Piao Engineering Research Center of Metallurgical Automation and Measurement Technology Ministry of Education Wuhan University of Science and Technology Wuhan Hubei 430081 China State Key Laboratory of Industrial Control Technology Zhejiang University Hangzhou 310027 China Department of Electronics Information and System Engineering Hanyang University 425-791 Korea Department of Electronic Information Engineering College of Science and Engineering Yanbian University Yanji 133002 China

Home automation systems based on wireless sensor/actuator networks are characterized by diversity of node power sources, limited computational power, and mobility. We propose a routing protocol that fully uses the location of static nodes to limit the search for a route to a small zone based on the analysis of characteristics and requirements. The calculation of two different kinds of zone and the route discovery procedure are described in this paper. According to the environment of home automation, we compared the performance of our routing protocols with the ad-hoc on-demand distance-vector protocol using simulation and gave suggestions for zone selection. Simulation results showed that our routing protocol dramatically reduced the routing overhead and increased the reliability.

关键词： Wireless Sensor/Actuator Networks Home automation Routing Protocol

来源：评论

学校读者我要写书评

暂无评论

一种基于直方图规定化的相干体增强方法

一种基于直方图规定化的相干体增强方法

引用

CPS/SEG北京2009国际地球物理会议

作者： WangJi Lu Wenkai State Key Laboratory of Intelligent Technology and Systems Tsinghua National Laboratory for Information Science and TechnologyDepartment of AutomationTsinghua UniversityBeijing City100084China

在地震资料解释中，相干体是用于描述地层的非连续性以寻找断层、裂缝等地质现象的有效工具。然而，规模较小的断层和微裂缝对应的相邻地震道之间的差异较小，同时又受到噪声的干扰，从而使其相干值较高。因此在相干体的时间切片上，难... 详细信息

在地震资料解释中，相干体是用于描述地层的非连续性以寻找断层、裂缝等地质现象的有效工具。然而，规模较小的断层和微裂缝对应的相邻地震道之间的差异较小，同时又受到噪声的干扰，从而使其相干值较高。因此在相干体的时间切片上，难于识别小断层和微裂缝。本文提出了一种基于直方图规定化的方法，对相干体中进行增强。对实际数据的处理结果表明，本文所提算法能有效增强相干体中的细节信息。

关键词：相干体断层识别直方图均衡直方图规定化地震资料解释小断层识别

来源：评论

学校读者我要写书评

暂无评论

Robust Controller Design for Networked Control systems with Nonlinear Uncertainties

Robust Controller Design for Networked Control Systems with ...

引用

2009 American Control Conference (ACC 2009)

作者： Junyan Yu Long Wang Mei Yu Jie Chen Yingmin Jia Intelligent Control Laboratory Center for Systems and Control College of Engineering and Key Laboratory of Machine Perception (Ministry of Education) Peking University Beijing 100871 P.R.China Department of Automation North China Electric Power University Zhuxinzhuang Dewai Beijing 102206 P.R.China Beijing Institute of Technology Key Laboratory of Complex System Intelligent Control and Decision Ministry of Education Beijing 100081 P.R.China Seventh Research Division Beihang University Beijing 100083 P.R. China

ISBN: (纸本)9781424445233

We address robust stabilization problem for networked control systems with nonlinear uncertainties and packet losses by modelling such systems as a class of uncertain switched systems. Based on theories on switched Lyapunov functions, we derive the robustly stabilizing conditions for state feedback stabilization and design packet-loss dependent controllers by solving some matrix inequalities. A numerical example and some simulations are worked out to demonstrate the effectiveness of the proposed design method.

关键词： Networked control systems robust controller design Packet loss control systems design approach Switched systems controller design state feedback Uncertainty nonlinearity

来源：评论

学校读者我要写书评

暂无评论

New method for interactive TCP traffic replay based on balance-checking between transmitted and received packets

引用

Jisuanji Xuebao/Chinese Journal of Computers 2009年第4期32卷 835-846页

作者： Chu, Wei-Bo Cai, Zhong-Min Guan, Xiao-Hong Chen, Ming-Xu MOE Key Laboratory for Intelligent Networks and Network Security Xi'an Jiaotong University Xi'an 710049 China State Key Laboratory for Manufacturing Systems Xi'an Jiaotong University Xi'an 710049 China Tsinghua National Laboratory for Information Science and Technology Department of Automation Tsinghua University Beijing 100084 China

Interactive network traffic replay is the newest method for testing and evaluation of network devices such as Firewalls, IPSes, routers, switches, etc. Currently state-checking method is used for interactive TCP traffic replay. This paper proposes a new method for interactive TCP traffic replay which is based on the balance status between transmitted and received packets. By checking the balance conditions before sending out TCP packets, the method can significantly reduce the cost of state-checking and enhance the replay performance. The authors made a comparison on the differences of replay methods when introducing the balance mechanism. The efficiency of the method is also investigated and evaluated from aspects of a single TCP session, multi-session traffic, packet losses and latency. Experimental results show that the method outperforms the original state-checking method when replaying actual TCP traffics.

关键词： Network security

来源：评论

学校读者我要写书评

暂无评论

Probabilistic support vector machine output adjusting for sampling bias

引用

Journal of Harbin Institute of technology (New Series) 2008年第SUPPL.期15卷 157-161页

作者： Yang, Chunyu Zhou, Jie State Key Laboratory on Intelligent Technology and Systems Department of Automation Tsinghua University Beijing 100084 China Tsinghua National Laboratory for Information Science and Technology Department of Automation Tsinghua University Beijing 100084 China

In some real-world classification tasks, the classifier may be trained on a data set which does not reflect the class distribution of the real data set. Such sampling bias or virtual concept drift may seriously affect the classification accuracy. Previous researches on this topic mainly concern classifiers with explicit a posteriori probabilities output. There has been a framework to adjust the original classifier using Expectation Maximization (EM) algorithm for such classifiers. The margin based classifier Support Vector Machine (SVM), has not been studied under this framework because of the lack of probabilistic output. In this paper, we discuss the probabilistic output of SVM and propose a Gaussian Mixture Model (GMM) to approximate the class conditional distribution of the margin so as to adjust the classifier using the EM framework. Experimental results on standard machine learning data sets show that the proposed algorithm can improve the classification accuracy on most of these problems. It performs especially well on those data sets with low classification accuracy.

关键词： Classification (of information)

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：