检索结果-内蒙古大学图书馆

IEEE National Conference on Aerospace and Electronics (NAECON)

作者： C.M. Hopper Yi Pan Wright Laboratories OH USA Computer Science Dcpmnent University of Dayton Dayton OH USA

Many avionics systems use specialized parallel architectures to speed processing and to increase system reliability. The software used therein is frequently divided into tasks and executed concurrently on multiple processors under strict real-time constraints critical to the mission's successful performance. Scheduling and planning are needed for effectively managing the computational resources on such avionics architectures. Since most real-time scheduling problems are known to be NP-hard, an approximation approach that applies heuristic methods using conventional computer algorithms has been used to solve these scheduling problems. Artificial intelligence (AI) planners have been used extensively in manufacturing scheduling and operations research. In this paper, we demonstrate the idea of using AI planners to perform scheduling through an example. We derive a solution to scheduling several image tasks on a distributed computer system, using the AI planner PRODIGY. The basic characteristics of AI planners in general and the PRODIGY solver in particular are described, the domain theory and problem specification for our problem through the PRODIGY description language PDL are presented.

关键词： distributed computing Artificial intelligence Job shop scheduling Processor scheduling Aerospace electronics Concurrent computing parallel architectures Reliability Software performance Resource management

来源：评论

学校读者我要写书评

暂无评论

distributed image edge detection methods and performance

Distributed image edge detection methods and performance

引用

Proceeedings of the 6th IEEE Symposium on parallel and distributed processing

作者： Zhang, Xiaodong Deng, Hong Univ of Texas at San Antonio San Antonio United States

An edge detection process in computer vision and image processing detects any types of significant features appearing as discontinuities in intensities. This paper presents our experience with parallelizing an edge detection application algorithm that reduces noise and unnecessary detail in a gray-scale image from a coarse level to a fine level of resolution by using an edge focusing technique. Numerical methods and parallel implementations of edge focusing are presented. The image detection algorithms are implemented on three representative massage-passing architectures: a low-cost heterogeneous PVM network, an Intel iPSC/860 hypercube, and a CM-5 massively parallel multicomputer. Our objectives are to provide insight into implementation and performance issues for image processing applications on general-purpose message-passing architectures, to investigate implications on network variations, and to evaluate the computing scalabilities on the three network systems by examining execution and communication patterns of the image edge detection application.

关键词： Edge detection

来源：评论

学校读者我要写书评

暂无评论

Texture analysis for image processing on general-purpose parallel machines

Texture analysis for image processing on general-purpose par...

引用

Proceedings of the International Symposium on parallel Architectures, Algorithms and Networks (ISPAN)

作者： Boroczky, Lilla Cremonesi, Paolo Scarabottolo, Nello Hungarian Acad of Sciences Budapest Hungary

The problem considered in this paper is the definition of an efficient parallel algorithm for texture analysis of an image. The target architectures are distributed-memory general-purpose MIMD parallel machine. The solutions here proposed are based on two different methods: the Statistic Feature Matrix and the Wavelet Decomposition.

关键词： image processing

来源：评论

学校读者我要写书评

暂无评论

parallel computation of 3D wavelets

Parallel computation of 3D wavelets

引用

Proceedings of the Scalable High-Performance Computing Conference

作者： Suzuki, Laura R.C. Reid, J.Robert Burns, Thomas J. Lamont, Gary B. Rogers, Steven K. Air Force Inst of Technology Wright-Patterson AFB United States

The Discrete Wavelet Transform (DWT) is becoming a widely used tool in image processing and other data analysis areas. A non-conventional variation of a spatio-temporal 3D DWT has been developed in order to analyze motion in time-sequential imagery. The computational complexity of this algorithm is Θ(n3), where n is the number of samples in each dimension of the input image sequence. methods are needed to increase the speed of these computations for large data sets. Fortunately, wavelet decomposition is very amenable to parallelization. Coarse-grained parallel versions of this process have been design and implemented on three different architectures: a distributed network represented by a distributed network of Sun SPARCstation 2 workstations;two Intel hypercubes (an iPSC/2 and an iPSC/860);and a Thinking Machines Corporation CM-5, a massively parallel SPMD. This non-conventional 3D wavelet decomposition is very suitable for course-grain implementation on parallel computers with proper load balancing. Close to linear speedup over serial implementations has been achieved using a distributed network. Near-linear speedup was obtained on the hypercubes and the CM-5 for a variety of image-processing applications.

关键词： parallel processing systems

来源：评论

学校读者我要写书评

暂无评论

Object-driven dataflow computing for image processing on heterogeneous distributed systems

Object-driven dataflow computing for image processing on het...

引用

作者： Lauro, Karen Lynn Wilken Case Western Reserve University

学位级别：Ph.D., Doctor of Philosophy, Ph.D./Interdisciplinary

Object-driven dataflow (ODF) is a methodology that provides a system to build applications on heterogeneous multiprocessing systems or networks that support multiple parallel processing topologies in a hierarchical structure. Its discriminating features are driven by the generalization of macro-dataflow computing to heterogeneous systems and the focus on large system development through object-oriented computing. Natural heterogeneous processing is examined to gain insight into the principles and conceptual models of heterogeneous processing which are integrated into ODF. ODF was developed to be applicable to both research and product development in the area of image processing. To enable this, it must provide load-time (not compile time) adaptability without sacrificing performance. The intent is to take advantage of the inherent parallelism in dataflow computing on high level, while allowing individual tasks to be developed with traditional imperative methods, if desired. The ODF determines the execution flow on a task level, in addition to providing the communication framework for intra-task communication. Dynamic binding is used to initiate the appropriate code on each of the nodes given the availability of the different types of nodes, object types, and the specific application. The objects are commonly decomposed to the grain-size appropriate to a given task on the available processing node.

关键词： Applied sciences

来源：评论

学校读者我要写书评

暂无评论

distributed image edge detection methods and performance

Distributed image edge detection methods and performance

引用

International Symposium on parallel and distributed processing (IPDPS)

作者： Xiaodong Zhang Hong Deng High-Performance Computing and Software Laboratory University of Texas San Antonio San Antonio TX USA

ISBN: (纸本)0818664274

An edge detection process in computer vision and image processing detects any types of significant features appearing as discontinuities in intensities. This paper presents our experience with parallelizing an edge detection application algorithm that reduces noise and unnecessary detail in a gray-scale image from a coarse level to a fine level of resolution by using an edge focusing technique. Numerical methods and parallel implementations of edge focusing are presented. The image detection algorithms are implemented on three representative message-passing architectures: a low-cost heterogeneous PVM network, an Intel iPSC/860 hypercube, and a CM-5 massively parallel multicomputer. Our objectives are to provide insight into implementation and performance issues for image processing applications on general-purpose message-passing architectures, to investigate implications an network variations, and to evaluate the computing scalabilities on the three network systems by examining execution and communication patterns of the image edge detection application.< >

关键词： image edge detection image processing Focusing Computer architecture Computer vision Application software Noise reduction Noise level Gray-scale image resolution

来源：评论

学校读者我要写书评

暂无评论

Texture analysis for image processing on general-purpose parallel machines

Texture analysis for image processing on general-purpose par...

引用

International Symposium on parallel Architectures, Algorithms and Networks (ISPAN)

作者： L. Boroczky P. Cremonesi N. Scarabottolo KFKI Research Institute for Measurement and Computing Techniques Hungarian Academy of Sciences (ATOMKI) Budapest Hungary Dipartimento di Elettronica e Informazione Politecnico di Milano Milan Italy

The problem considered in this paper is the definition of an efficient parallel algorithm for texture analysis of an image. The target architectures are distributed-memory general-purpose MIMD parallel machines. The solutions proposed here are based on two different methods, the Statistic Feature Matrix and the Wavelet Decomposition.< >

关键词： image analysis image texture analysis image processing parallel machines Concurrent computing Wavelet analysis Algorithm design and analysis distributed computing Power system interconnection Computer networks

来源：评论

学校读者我要写书评

暂无评论

parallel computation of 3D wavelets

Parallel computation of 3D wavelets

引用

Scalable High Performance Computing Conference

作者： L.R.C. Suzuki J.R. Reid T.J. Burns G.B. Lamont S.K. Rogers Graduate School of Engineering AFIT/ENG Air Force Institute of Technology Dayton OH USA

The Discrete Wavelet Transform (DWT) is becoming a widely used tool in image processing and other data analysis areas. A non-conventional variation of a spatiotemporal 3D DWT has been developed in order to analyze motion in time-sequential imagery. The computational complexity of this algorithm is /spl Theta/(n/sup 3/), where n is the number of samples in each dimension of the input image sequence. methods are needed to increase the speed of these computations for large data sets. Fortunately, wavelet decomposition is very amenable to parallelization. Coarse-grained parallel versions of this process have been designed and implemented on three different architectures: a distributed network represented by a distributed network of Sun SPARCstation 2 workstations: two Intel hypercubes (an iPSC/2 and an iPSC/860); and a Thinking Machines Corporation CM-5, a massively parallel SPMD. This non-conventional 3D wavelet decomposition is very suitable for coarse-grain implementation on parallel computers with proper load balancing. Close to linear speedup over serial implementations has been achieved using a distributed network. Near-linear speedup was obtained on the hypercubes and the CM-5 for a variety of image-processing applications.< >

关键词： Concurrent computing Discrete wavelet transforms Hypercubes Wavelet analysis image processing Data analysis Spatiotemporal phenomena image analysis image motion analysis Motion analysis

来源：评论

学校读者我要写书评

暂无评论

Communication and computation patterns of large scale image convolutions on parallel architectures

Communication and computation patterns of large scale image ...

引用

International Symposium on parallel processing

作者： S.G. Dykes Xiaodong Zhang Yan Zhou Haixu Yang High Performance Computing and Software Laboratory University of Texas San Antonio San Antonio TX USA

Segmentation and other image processing operations rely on convolution calculations with heavy computational and memory access demands. The article presents an analysis of a texture segmentation application containing a 96/spl times/96 convolution. Sequential execution required several hours an single processor systems with over 99% of the time spent performing the large convolution. 70% to 75% of execution time is attributable to cache misses within the convolution. We implemented the same application on CM-5, iPSC/860 and PVM distributed memory multicomputers, tailoring the parallel algorithms to each machine's architecture. parallelization significantly reduced execution time, taking 49 seconds on a 512 node CM-5 and 6.5 minutes on a 32 node iPSC/860. The results indicate for large kernel convolutions the size and bandwidth of the fast memory store is more important than processor power or communication overhead.< >

关键词： Concurrent computing Large-scale systems parallel architectures image segmentation image processing Kernel Laboratories Bandwidth Pixel Smoothing methods

来源：评论

学校读者我要写书评

暂无评论

基于逻辑神经网络的图象处理方法

引用

航空学报 1994年第5期15卷 570-575页

作者：杨国庆葛洪纬南京航空航天大学计算机系

阐述了基于逻辑神经网络的图象平滑、压缩编码方法。这些方法的基本思想是：以逻辑神经网络为窗口，选择对某种图象变换敏感的模式训练窗口。当窗口对图象扫描时，窗口在图象的各位置输出响应，通过对这些响应的简单运算，实现所希望的... 详细信息

阐述了基于逻辑神经网络的图象平滑、压缩编码方法。这些方法的基本思想是：以逻辑神经网络为窗口，选择对某种图象变换敏感的模式训练窗口。当窗口对图象扫描时，窗口在图象的各位置输出响应，通过对这些响应的简单运算，实现所希望的处理。除了理论分析外，文中还给出了各种方法的典型试验结果。

关键词：图象处理逻辑设计神经网络

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：