检索结果-内蒙古大学图书馆

Fast Optical Flow Estimation Without Parallel Architectures

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2017年第11期27卷 2322-2332页

作者： Zhu, En Li, Yuanwei Shi, Yanling Natl Univ Def Technol Sch Comp Sci Top Key Discipline Comp Software & Theory Zhejian Changsha 410073 Hunan Peoples R China Chongqing Commun Inst Chongqing Key Lab Emergency Commun Chongqing 400035 Peoples R China Cent South Univ Changsha 410083 Hunan Peoples R China

According to recent results on Middlebury, MPI Sintel, and KITTI benchmarks, the accuracy of optical flow estimation algorithms has been significantly improved. The speed of them, however, has been too slow to meet the requirement of real-time applications. As a result, some parallel architectures (such as FPGA or GPU) have to be used for accelerating. Therefore, reducing the computational cost of optical flow estimation makes a lot of sense. To overcome the above issues, this paper proposes a fast local method based on 3D-gradients and approximate nearest-neighbor field (NNF), which is different from the widely used global model. In our method, NNF is used to provide initial optical flow field, and the proposed fast 3D-gradients-based local operator is used to propagate flow from coarse level to finer level in the coarse-to-fine refinement. We implement two versions of our method (with/without NNF initialization). Experimental results show that our method has a significant advantage for speed over other methods, where the fast version is capable of processing Urban sequence (640x480) at approximate to 10 frames/s without parallel architectures. Meanwhile, our accuracy is also within acceptable levels on both small and large motion for some real-time applications.

关键词： 3D-gradient nearest-neighbor field (NNF) optical flow

来源：评论

学校读者我要写书评

暂无评论

Detecting and counting pedestrians in real time

引用

Journal of Computational Information Systems 2014年第2期10卷 827-835页

作者： Li, Yuanwei Zhu, En Zhao, Jianming Zhu, Xinzhong Yin, Jianping College of Computer National University of Defense Technology Changsha 410073 China Top Key Discipline of Computer Software and Theory in Zhejiang Provincial Colleges Zhejiang Normal University Jinhua 321004 China College of Mathematics Physics and Information Engineering Zhejiang Normal University Jin Hua 321004 China State Key Laboratory of High Performance Computing National University of Defense Technology Changsha 410073 China

In this paper, we present a real time system for estimating the number of pedestrians from detection results in crowd, which may be very large and dense. Firstly, the scene is segmented into foreground and background, and pedestrian detection is performed on foreground area with GPU. In order to detect occluded pedestrians effectively, two detectors are used to detect two parts of pedestrian (upper body and full body). Then, we fuse two kinds of detection results by a series of strategies to obtain the final detection results, which are expected to be better than any one of the two. Thirdly, the number of final detection results is regarded as the estimate of the number of pedestrians after smoothed by kalman filter. Experimental evaluation on PEST 2009 data set is given and shows that our system is real time and effective. Copyright © 2014 Binary Information Press.

关键词： Kalman filters

来源：评论

学校读者我要写书评

暂无评论

Counting Pedestrian with Mixed Features and Extreme Learning Machine

引用

COGNITIVE COMPUTATION 2014年第3期6卷 462-476页

作者： Li, Yuanwei Zhu, En Zhu, Xinzhong Yin, Jianping Zhao, Jianmin Natl Univ Def Technol Coll Comp Changsha 410073 Hunan Peoples R China Zhejiang Normal Univ Top Key Discipline Comp Software & Theory Zhejiang Prov Coll Jinhua 321004 Peoples R China Natl Univ Def Technol State Key Lab High Performance Comp Changsha 410073 Hunan Peoples R China

An approach for estimating the number of pedestrians is presented in this paper. The proposed counting framework combines two main pedestrian counting strategies-direct approach and indirect approach-by the use of mixed features and extreme learning machine (ELM). ELM is used to map mixed features to the number of pedestrians. Mixed features consist of holistic low-level features and rectangular local binary pattern (rLBP) features, and rLBP features are new features designed to describe the statistical and structural information of explicit pedestrian detection rectangles. Through mixed features, the information from both direct approach (rLBP features) and indirect approach (low-level features) is used in our algorithm, so we can take full advantage of two counting strategies. The detection rectangles are obtained by the use of the pedestrian detector described in paper "the fastest pedestrian detector in the west" (FPDW) by Dollar et al. Based on integral channel features and soft cascade classifier, FPDW is able to provide outstanding detection results at rapid speed. Experimental results on PETS 2009 datasets show that the proposed counting framework can improve counting accuracy significantly by the combination of two counting strategies. rLBP features are effective to describe the useful information of detection rectangles for regression models, and mixed features are more effective than either of both.

关键词： Extreme learning machine Gaussian processes Pedestrian detection Pedestrian counting Rectangular local binary pattern Kalman filter

来源：评论

学校读者我要写书评

暂无评论

Deriving Requirements Specification With Time: A software Environment Ontology based Approach

Deriving Requirements Specification With Time: A Software En...

引用

IEEE 37th Annual computer software and Applications Conference (COMPSAC)

作者： Chen, Xiaohong Ye, Ronghua Sun, Haiying Lu, Hong East China Normal Univ Shanghai Key Lab Trustworthy Comp Shanghai Peoples R China Zhejiang Normal Univ Coll Math Phys & Informat Engn Jinhua Zhejiang Peoples R China Zhejiang Normal Univ Top Key Discipline Comp Software & Theory Zhejian Jinhua Zhejiang Peoples R China

ISBN: (纸本)9780769549866

It is well acknowledged that environment plays an important role in requirement derivation. However, at present the time-continuous properties of the environment are of little concern. Our previous work modeled the time-continuous environment by constructing a software environment ontology. This paper further presents an approach for deriving software requirements specification with time using software environment ontology. Experiments are conducted by deriving different software requirement specifications under different situations. They are simulated with Simulink. The simulation results show that the software behaviors can be more accurately determined with respect to time-continuous environment by using time as a measurement.

关键词： requirements engineering environment time-continuous property software environment ontology requirement specification derivation

来源：评论

学校读者我要写书评

暂无评论

26th Annual Computational Neuroscience Meeting (CNS*2017): Part 3 Antwerp, Belgium. 15-20 July 2017 Abstracts

引用

BMC NEUROSCIENCE 2017年第SUPPL 1期18卷 95-176页

作者： [Anonymous] Department of Neuroscience Yale University New Haven CT 06520 USA Department Physiology & Pharmacology SUNY Downstate Brooklyn NY 11203 USA NYU School of Engineering 6 MetroTech Center Brooklyn NY 11201 USA Departament de Matemàtica Aplicada Universitat Politècnica de Catalunya Barcelona 08028 Spain Institut de Neurobiologie de la Méditerrannée (INMED) INSERM UMR901 Aix-Marseille Univ Marseille France Center of Neural Science New York University New York NY USA Aix-Marseille Univ INSERM INS Inst Neurosci Syst Marseille France Laboratoire de Physique Théorique et Modélisation CNRS UMR 8089 Université de Cergy-Pontoise 95300 Cergy-Pontoise Cedex France Department of Mathematics and Computer Science ENSAT Abdelmalek Essaadi’s University Tangier Morocco Laboratory of Natural Computation Department of Information and Electrical Engineering and Applied Mathematics University of Salerno 84084 Fisciano SA Italy Department of Medicine University of Salerno 84083 Lancusi SA Italy Dipartimento di Fisica Università degli Studi Aldo Moro Bari and INFN Sezione Di Bari Italy Data Analysis Department Ghent University Ghent Belgium Coma Science Group University of Liège Liège Belgium Cruces Hospital and Ikerbasque Research Center Bilbao Spain BIOtech Department of Industrial Engineering University of Trento and IRCS-PAT FBK 38010 Trento Italy Department of Data Analysis Ghent University Ghent 9000 Belgium The Wellcome Trust Centre for Neuroimaging University College London London WC1N 3BG UK Department of Electronic Engineering NED University of Engineering and Technology Karachi Pakistan Blue Brain Project École Polytechnique Fédérale de Lausanne Lausanne Switzerland Departement of Mathematics Swansea University Swansea Wales UK Laboratory for Topology and Neuroscience at the Brain Mind Institute École polytechnique fédérale de Lausanne Lausanne Switzerland Institute of Mathematics University of Aberdeen Aberdeen Scotland UK Department of Integrativ

来源：评论

学校读者我要写书评

暂无评论

Problem Frames Construction from Feature Models

Problem Frames Construction from Feature Models

引用

Asia-Pacific Conference on software Engineering

作者： Xiaohong Chen Haiying Sun Ronghua Ye Jing Liu Shanghai Key Laboratory of Trustworthy Computing East China Normal University Shanghai China Top Key Discipline of Computer Software and Theory in Zhejiang Provincial Colleges Zhejiang Normal University

ISBN: (纸本)9781479921454

The Problem Frames (PF) approach is a well-known approach for describing, analyzing and structuring problems in requirements engineering. It defines certain patterns of problems to be problem frames which have solutions. The real world problems are solved by decomposing to sub-problems which could be matched against problem frames. However, whether existing problem frames is enough for covering all problems is undecided. In this paper, we propose to recognize problem frames in a specific application domain by using feature models. As feature models could link to solution space, our approach bridges the gap between problem frames and solution space. The newly constructed problem frames could be used to analyse problems. We illustrate their usage by presenting a problem frames based problem analysis which include problem descriptions and matches. Because our new problem frames are high level, the real world problems do not need to be decomposed before matched.

关键词： Wind speed Ocean temperature Temperature measurement Temperature sensors Sea measurements Educational institutions

来源：评论

学校读者我要写书评

暂无评论

A Multimodal Fusion Algorithm Based on FRR and FAR Using SVM

引用

INTERNATIONAL JOURNAL OF SECURITY AND ITS APPLICATIONS 2013年第3期7卷 65-74页

作者： Li, Yong Shit, Meimei Zhu, En Yin, Jianping Zhao, Jianmin Engn Univ CAPF Dept Informat Engn Xian Shaanxi Peoples R China Engn Univ CAPF Dept Sci Xian Shaanxi Peoples R China Natl Univ Def Technol Sch Comp Changsha Hunan Peoples R China Zhejiang Normal Univ Zhejiang Prov Colleges Top Key Discipline Comp Software & Theory Jinhua Zhejiang Peoples R China

Remarkable improvements in recognition can be achieved through multibiometric fusion. Among various fusion techniques, score level fusion is the most frequently used in multibiometric system. In this paper, we propose a novel fusion algorithm based on False Reject Rate (FRR) and False Accept Rate (FAR) using Support Vector Machine (SVM). It transfers scores into corresponding FRRs and FARs, thus avoiding calculating posteriori probability of a certain score, as well as be capable of illustrating distribution of matching scores. The proposed method takes full advantages of both capabilities of FRR and FAR to describe the order of score and classification of SVM. Experimental results show that the proposed method outperforms existing representative approaches and can effectively improve the performance of multibiometric system.

关键词： Biometrics Multibiometric Score level fusion Multi modal SVM False Reject Rate False Accept Rate

来源：评论

学校读者我要写书评

暂无评论

Problem Frames Construction from Feature Models 20

Problem Frames Construction from Feature Models

引用

20th Asia-Pacific software Engineering Conference (APSEC)

作者： Chen, Xiaohong Sun, Haiying Ye, Ronghua Liu, Jing E China Normal Univ Shanghai Key Lab Trustworthy Comp Shanghai 200062 Peoples R China Zhejiang Normal Univ Coll Math Phys & Informat Engn Jinhua City Zhejiang Peoples R China Zhejiang Normal Univ Top Key Discipline Comp Software & Theory Zhejian Jinhua City Zhejiang Peoples R China

ISBN: (纸本)9781479921430

The Problem Frames (PF) approach is a well-known approach for describing, analyzing and structuring problems in requirements engineering. It defines certain patterns of problems to be problem frames which have solutions. The real world problems are solved by decomposing to sub-poblems which could be matched against problem frames. However, whether existing problem frames is enough for covering all problems is undecided. In this paper, we propose to recognize problem frames in a specific application domain by using feature models. As feature models could link to solution space, our approach bridges the gap between problem frames and solution space. The newly constructed problem frames could be used to analyse problems. We illustrate their usage by presenting a problem frames based problem analysis which include problem descriptions and matches. Because our new problem frames are high level, the real world problems do not need to be decomposed before matched.

关键词： Requirements Engineering Problem Frames Environment Ontology Problem Analysis Feature Model

来源：评论

学校读者我要写书评

暂无评论

Development of visual action design environment for intelligent toy robot

Development of visual action design environment for intellig...

引用

3rd International Conference on Swarm Intelligence, ICSI 2012

作者： Mo, Jianqing He, Hanwu Zhang, Hong Guangdong University of Technology Guangzhou China Top Key Discipline of Computer Software and Theory in Zhejiang Provincial Colleges Zhejiang Normal University Jinhua China Guangdong South China Institute of Industrial Design Dongguan China

ISBN: (纸本)9783642309755

To improve efficiency of action design for intelligent toy robot, a visual action design environment is created. Compared with the text-based action design software, it's intuitive, good usability. An algorithm for an automatic generation action sequence file (ASF) is present according to the data structure of action sequence, the file structure of ASF and the judgment methods of operation rules. Then using Eon Studio as virtual simulation development platform, the analysis to complicated action sequence files and reappearance of virtual action are realized. © 2012 Springer-Verlag.

关键词： Toys

来源：评论

学校读者我要写书评

暂无评论

Voronoi diagram with visual restriction

Voronoi diagram with visual restriction

引用

6th International Frontiers of Algorithmics Workshop, FAW 2012 and 8th International Conference on Algorithmic Aspects of Information and Management, AAIM 2012

作者： Fan, Chenglin Luo, Jun Wang, Wencheng Zhu, Binhai Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences China Institute of Software Chinese Academy of Sciences China Top Key Discipline of Computer Software and Theory Zhejiang Provincial Colleges Zhejiang Normal University China Department of Computer Science Montana State University Bozeman MT 59717 United States

ISBN: (纸本)9783642296994

In a normal Voronoi diagram, each site is able to see all the points in the plane. In this paper, we study the case such that each site is only able to see a visually restricted region in the plane and construct the so-called Visual Restriction Voronoi Diagram (VRVD). We show that the visual restriction Voronoi cell of each site is not necessary convex and it could consist of many disjoint regions. We prove that the combinatorial complexity of the VRVD on n sites is Θ(n 2). Then we give an O(n 2logn) time and O(n 2) space algorithm to construct VRVD. © 2012 Springer-Verlag.

关键词： Computational geometry

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：