检索结果-内蒙古大学图书馆

International Conference on Advanced Information Networking and Applications (AINA)

作者： C. Gammage V. Chaudhary Computer Science Department Wayne State University Detroit MI USA Institute for Scientific Computing Wayne State University Detroit MI USA

Fuzzy Connectedness is an important image segmentation routine for image processing of medical images. It is often used in preparation for surgery and sometimes during surgery. It is important to have an algorithm which can execute very fast, especially in the intra-operative environment. We have taken code from a popular image processing toolkit called ITK and ported it to a C environment. We optimized the implementation to give maximal performance (giving speedup of 23 times). We attempted three different levels of parallelization. We found that MPI was not an efficient method of parallelization as the algorithm is data dependant and large amounts of communication must be done. This communication overshadows the speed increase from doing computation on multiple processors, or nodes in a cluster. However, some limited speedup over the optimizations was obtained using OpenMP on an SMP system leading to a speedup of fifty using four processors over the original ITK implementation.

关键词： Image segmentation Biomedical imaging Magnetic resonance imaging Fuzzy sets Surgery Image processing Computed tomography Clustering algorithms Positron emission tomography computer science

来源：评论

学校读者我要写书评

暂无评论

On the Advantages of Weighted L1-Norm Support Vector Learning for Unbalanced Binary Classification Problems

On the Advantages of Weighted L1-Norm Support Vector Learnin...

引用

International IEEE Conference on Intelligent Systems, IS

作者： Tatjana Eitrich Bruno Lang Central Institute for Applied Mathematics Research Center Juelich GmbH Germany Applied Computer Science and Scientific Computing Group Department of Mathematics University of Wuppertal Germany

In this paper we analyze support vector machine classification using the soft margin approach that allows for errors and margin violations during the training stage. Two models for learning the separating hyperplane do exist. We study the behavior of the optimization algorithms in terms of training characteristics and test accuracy for unbalanced data sets. The main goal of our work is to compare the features of the resulting classification functions, which are mainly defined by the support vectors arising during the support vector machine training

关键词： Support vector machine classification Support vector machines Learning systems Kernel Testing Supervised learning Mathematics Intelligent systems Machine learning Machine learning algorithms

来源：评论

学校读者我要写书评

暂无评论

Internet-Based Collaborative Virtual Simulations with Area of Interest Management

Internet-Based Collaborative Virtual Simulations with Area o...

引用

International Symposium on Collaborative Technologies and Systems

作者： A. Diabi S. Shirmohammadi A. Gillmore P. Lacombe J.C. de Oliveira Distributed Collaborative Virtual Environments Research Laboratory (DISCOVER Lab) University of Ottawa Ottawa Canada Computer Science Department National Laboratory for Scientific Computing Petropolis Brazil

IP multicast is typically used to support message transmission among entities in collaborative virtual simulations running on Intranets, specially military training simulations. But, IP Multicast is mostly not available on the Internet. Alternatively, researches have in recent years proposed the use of Application Layer Multicasting techniques (ALM) to alleviate this problem and to allow a somewhat scalable message passing among peers in a group of users on the Internet. As the number of users grows in the session, efficient handling of resources and scalable communication between users becomes critical due to network traffic and computation required to track all objects. The solution typically used is to partition the virtual simulation into a number of areas, and to manage each area separately. This is known as Area of Interest Management (AoIM). In this paper, we propose a peer-to-peer communication architecture for collaborative simulations along with a suitable AoIM solution. Our architecture uses both proxies and endsystems to provide a number of communication services suitable for Intranet, Internet, and hybrid-based collaborative virtual simulations.

关键词： Internet Collaboration Computational modeling Management training Message passing Telecommunication traffic Traffic control computer networks Peer to peer computing computer architecture

来源：评论

学校读者我要写书评

暂无评论

An Efficient Commit Protocol Exploiting Primary-Backup Placement in a Distributed Storage System

An Efficient Commit Protocol Exploiting Primary-Backup Place...

引用

Pacific Rim International Symposium on Dependable computing

作者： Xiangyong Ouyang Tomohiro Yoshiharay Haruo Yokota Department of Computer Science Tokyo Institute of Technology Meguro-ku Tokyo Japan Global Scientific Information and Computing Center Tokyo Institute of Technology Meguro-ku Tokyo Japan

Advanced data engineering applications require a large-scale storage system that is both scalable and dependable. In such a system, an atomic commit protocol becomes imperative to ensure the consistency and atomicity of transactions. In this paper we present a new commit protocol, BA-1.5PC, which is well tailored to such distributed storage environments as autonomous disks that use a primary-backup storage schema. The protocol achieves an efficient commit process while also guaranteeing a high dependability by combining several approaches: (1) a low-overhead log mechanism that eliminates blocking disk I/Os, (2) removing the voting phase from commit processing to gain a faster commit process, and (3) a primary-backup assisted recovery strategy to enhance dependability in the presence of possible failures, so that a master failure in the decision phase are not block prepared cohorts of a transaction. Experiments were carried out on a trial version of an autonomous disks system to verify its efficiency. The results indicate that this protocol significantly outperforms several well-known commit protocols in terms of transaction throughput

关键词： Voting Access protocols Data engineering Application software Throughput Degradation computer science Information science Distributed computing Disk drives

来源：评论

学校读者我要写书评

暂无评论

A concurrency control method for parallel btree structures 22

A concurrency control method for parallel btree structures

引用

22nd International Conference on Data Engineering Workshops, ICDEW 2006

作者： Yoshihara, Tomohiro Kobayashi, Dai Taguchi, Ryo Yokota, Haruo Department of Computer Science Tokyo Institute of Technology Graduate School of Information Science and Engineering 2-12-1 Ookayama Meguro-ku Tokyo152-8552 Japan Science and Technical Research Laboratories Japan Broadcasting Corporation 1-10-11 Kinuta Setagaya-ku Tokyo157-8510 Japan Global Scientific Information and Computing Center Tokyo Institute of Technology 2-12-1 Ookayama Meguro-ku Tokyo152-8550 Japan

ISBN: (纸本)0769525717

A new concurrency control protocol for parallel Btree structures, MARK-OPT, is proposed. MARK-OPT marks the lowest structure-modification-operation (SMO) occurrence point during optimistic latch coupling operations, to reduce the cost of SMO compared to the conventional protocols such as ARIES/IM and INC-OPT. The marking reduces the frequency of restarts for spreading the range of X latches, which will clearly improves the system throughput. Moreover, the MARK-OPT is deadlock free and satisfies the physical consistency requirement for Btrees. These indicate that the MARK-OPT is right and suitable as a concurrency control protocol for Btree structures. This paper also proposes three variations of the protocol, INC-MARKOPT, 2P-INT-MARK-OPT and 2P-REP-MARK-OPT, by focusing on tree structure changes from other transactions. We implement the proposed protocols, the INC-OPT, and the ARIES/IM for the Fat-Btree, a form of parallel Btree, and compare the performance of these protocols using a large-scale blade system. The experimental results indicate that the proposed protocols always improve the system throughput, and the 2P-REP-MARK-OPT is the most useful protocol in a high update environment. Moreover, in the experiment, the low frequency of restarts in the proposed protocols indicates that the marking in the proposed protocols is effective. © 2006 IEEE.

关键词： Concurrency control

来源：评论

学校读者我要写书评

暂无评论

A Novel Method for Supporting Massively Multi-user Virtual Environments

A Novel Method for Supporting Massively Multi-user Virtual E...

引用

IEEE International Workshop on Haptic, Audio and Visual Environments and Their Applications

作者： Dewan Tanvir Ahmed Shervin Shirmohammadi Jauvane C. De Oliveira Distributed & Collaborative Virtual Environments Research Laboratory (DISCOVER Lab) University of Ottawa Ottawa Ontario Canada Computer Science Department National Laboratory for Scientific Computing Petrópolis Brazil

In collaborative distributed virtual environment people interact with each other to shard their states. In this paper we present massively multi-user virtual simulation architecture - MMVISA. The framework partitions the simulation platform into multiple regions to properly organize the decorative entities and to efficiently manage their association. The coordinator, the leader of a zone, manages local communications in multicast fashion but this multicast functionality is shifted from the network layer to the application layer to get the benefit of the scalability and easy deployability. Considering the behavior of the entities, coordinator opens multiple multicast channels to reduce structural reformation events among the entities. On the other hand, coordinators themselves form a top level mesh hierarchy to manage the area of interest. A mathematical model is given to determine the best node in the best zone with a given interest vector. This gives a provision to a node to import all the interesting messages when needed and makes it easy to be virtually there

关键词： Virtual environment Peer to peer computing Scalability Virtual reality Conferences Haptic interfaces Application software Collaborative work International collaboration Information technology

来源：评论

学校读者我要写书评

暂无评论

Dynamic contaminant identification in water

引用

ICCS 2006: 6th International Conference on Computational science

作者： Douglas, Craig C. Clay Harris, J. Iskandarani, Mohamed Johnson, Chris R. Lodder, Robert J. Parker, Steven G. Cole, Martin J. Ewing, Richard Efendiev, Yalchin Lazarov, Raytcho Qin, Guan University of Kentucky Department of Computer Science 773 Anderson Hall Lexington KY 40506-0046 United States Yale University Department of Computer Science P.O. Box 208285 New Haven CT 06520-8285 United States University of Kentucky Department of Chemistry Lexington KY 40506 United States University of Miami Rosenstiel School of Marine and Atmospheric Science 4600 Rickenbacker Causeway Miami FL 33149-1098 United States University of Utah Scientific Computing and Imaging Institute Salt Lake City UT 84112 United States Texas AandM University Institute for Scientific Computation 3404 TAMU 612 Blocker College Station TX 77843-3404 United States

ISBN: (纸本)3540343830

We describe how we plan to convert a traditional data collection sensor and ocean model into a DDDAS enabled system for identifying contaminants and then reacting with different models, simulations, and sensing strategies in a symbiotic manner. The sensor is just as useful in water as it would be on Mars for material identification. A successful terrestrial application of the sensor will lead to many new applications of the device and possible technology transfer to the private sector. © Springer-Verlag Berlin Heidelberg 2006.

关键词： Water pollution

来源：评论

学校读者我要写书评

暂无评论

A high-availability software update method for distributed storage systems

引用

Systems and computers in Japan 2006年第10期37卷 35-46页

作者： Kobayashi, Dai Watanabe, Akitsugu Uehara, Toshihiro Yokota, Haruo Department of Computer Science Graduate School of Information Science and Engineering Tokyo Institute of Technology Tokyo 152-8552 Japan NHK Science and Technical Research Laboratories Tokyo 157-8510 Japan Global Scientific Information and Computing Center Tokyo Institute of Technology Tokyo 152-8552 Japan

In this paper, we propose a nonstop system upgrade method without significant performance degradation for data management software. To reduce disk accesses and network traffic, we construct logical nodes inside a physical node and migrate data between the symbiotic logical nodes. This logical migration is assisted by storage management functions that hide data location and migration from users. We also show the effectiveness of our method using experimental results on the system based on the Autonomous Disks we have proposed as a highly available storage system technology. © 2006 Wiley Periodicals, Inc.

关键词： computer software

来源：评论

学校读者我要写书评

暂无评论

Treatment of Laser Pointer and Speech Information in Lecture Scene Retrieval

Treatment of Laser Pointer and Speech Information in Lecture...

引用

IEEE International Symposium on Multimedia

作者： Wataru Nakano Takashi Kobayashi Yutaka Katsuyama Satoshi Naoi Haruo Yokota Department of Computer Science Tokyo Institute of Technology Japan Global Scientific Information and Computing Center Tokyo Institute of Technology Japan Fujitsu Laboratories Limited Japan PUJITSU LABORATORIES Ltd.

ISBN: (纸本)0769527469;9780769527468

We have previously proposed a unified presentation contents search mechanism named UPRISE (unified presentation slide retrieval by impression search engine), and have also proposed a method to use laser pointer information in lecture scene retrieval. In this paper, we discuss the treatment of the laser pointer and speech information, and propose two methods to filter the laser pointer information using keyword occurrence in slides and speech. We also propose weighting schemata with filtered laser pointer information using slide text and speech information. We evaluate our approach by using actual lecture videos and presentation slides

关键词： Speech Layout Information retrieval Content based retrieval Information filtering Information filters Laser modes Laser theory Optical materials Search engines

来源：评论

学校读者我要写书评

暂无评论

Annotating user-defined abstractions for optimization 06

Annotating user-defined abstractions for optimization

引用

International Symposium on Parallel and Distributed Processing (IPDPS)

作者： D. Quinlan M. Schordan R. Vuduc Qing Yi Lawrence Livermore National Laboratory Center for Applied Scientific Computing Livermore CA USA Institute of Computer Languages University of Technology Vienna Vienna Austria Department of Computer Science University of Texas San Antonio San Antonio TX USA

ISBN: (纸本)9781424400546

Although conventional compilers implement a wide range of optimization techniques, they frequently miss opportunities to optimize the use of abstractions, largely because they are not designed to recognize and use the relevant semantic information about such abstractions. In this position paper, we propose a set of annotations to help communicate high-level semantic information about abstractions to the compiler, thereby enabling the large body of traditional compiler optimizations to be applied to the use of those abstractions. Our annotations explicitly describe properties of abstractions that are needed to guarantee the applicability and profitability of a broad variety of such optimizations, including memoization, reordering, data layout transformations, and inlining and specialization

关键词： Optimizing compilers Design optimization scientific computing Laboratories Profitability computer languages computer science Costs US Department of Energy Containers

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：