检索结果-内蒙古大学图书馆

iee colloquium on Practical applications of parallel Signal processing

作者： P. Dzwig Parsys Limited Hayes NJ USA

The Parsys SN 1000 series of parallel processing computer systems is considered as a vehicle for the generation of complex scenes using techniques such as ray-tracing and radiosity. The use of high throughput data cap... 详细信息

关键词： image processing parallel processing

来源：评论

学校读者我要写书评

暂无评论

Stereo matching of satellite images using the transputer

Stereo matching of satellite images using the transputer

引用

iee colloquium on Practical applications of parallel Signal processing

作者： K.A. Collins J.B.G. Roberts Central Research Laboratories THORN EMI Hayes Middlesex UK Scottish Power Public Limited Company Malvern Worcestershire UK

The transputer provides a powerful new capacity in parallel computing which can readily be optimised for specific applications. The authors compare three alternative decomposition techniques for a large problem on a network of 21 transputers, from the standpoints of speed, hardware efficiency and ease of programming. They have programmed three highly parallel implementations of an algorithm which constitutes part to a real-time system to generate terrain relief maps from satellite stereo image pairs. The particular algorithm used is the stereo matching strategy due to Barnard and Thompson (1980). The algorithm attempts to identify corresponding points in two image frames, in this case raw (non-epipolar) SPOT images, in order to refine initial estimates for the relative orientation parameters of the sensor in the two cases.< >

关键词： image processing parallel algorithms

来源：评论

学校读者我要写书评

暂无评论

An automated approach to mapping DSP algorithms onto transputer arrays

An automated approach to mapping DSP algorithms onto transpu...

引用

iee colloquium on Digital Signal processing for VLSI

作者： A.J. Goddard S.S. Lawson Centre for Information Engineering City University UK Department of Engineering Warwick University UK

There is an increasing demand for low cost, high throughput multiprocessors on which to implement real-time DSP applications. A general purpose solution can be offered using an MIMD architecture. MIMD architectures can come in a variety of different forms. Loosely-coupled, homogeneous architectures are examined, because they offer infinite expandability and a low cost:processing ratio. One potential disadvantage of loosely-coupled architectures arises at run-time, when program and data have to be dynamically scheduled onto processors. Run-time scheduling overheads tend to curtail performance. However, many real-time DSP applications have a deterministic structure. This allows scheduling to be performed at compile-time, in a static fashion, and so reduce run-time overheads. Static scheduling involves mapping the application (in the form of a task graph) onto the processor architecture (in the form of a processor graph).< >

关键词： parallel architectures Microprocessors

来源：评论

学校读者我要写书评

暂无评论

Cyclic architectures for concurrent processing applications

Cyclic architectures for concurrent processing applications

引用

iee colloquium on Practical applications of parallel Signal processing

作者： R.G. Bramley D.J. Creasey School of Electronic and Electrical Engineering University of Binningham Birmingham UK

An optimal algorithm could be defined as one that exploits the properties of the available technology to the full. That is, it takes account of the inadequacies in an implementation and exploits the merits. For example, when multiplication was a very slow operation, algorithms aimed to reduce the number of multiplications needed. The radix-2 fast Fourier transform is an example of this type of algorithm. Similarly, number-theoretic transforms sought to replace multiplications with shifts and additions: operations that could be performed much faster within the constraints of the then existing technology. The advent of VLSI technology has created a fundamental change in the meaning of algorithm optimisation. In VLSI systems, global communication is very difficult but multiplications and additions are fast. The structure of the computation is the most important factor determining throughput. A simple computational structure is demanded. Algorithms must exhibit recursiveness and replication and they must require only local communications for a good VLSI implementation.< >

关键词： Matrices Very-large-scale integration

来源：评论

学校读者我要写书评

暂无评论

A WSI architecture for AI semantic networks

A WSI architecture for AI semantic networks

引用

iee colloquium on VLSI for parallel processing

作者： J.G. Delgado-Frias W.R. Moore Department of Engineering Science University of Oxford UK

Many artificial intelligence (AI) applications need high performance computers that allow them to execute algorithms at reasonable speed. In recent years, researchers at universities and industry have developed new architectures in order to accomplish the performance required. A number of the AI algorithms require knowledge representation and manipulation which can be obtained by means of semantic networks. A WSI architecture dedicated to semantic network processing may have an impact on the AI field; since the parallelism in this application can be exploited by means WSI technology which offers a potential of higher reliability and speed. However, it is necessary to deal with some of the WSI problems, mainly defects on the silicon.< >

关键词： Artificial intelligence parallel architectures Very-large-scale integration

来源：评论

学校读者我要写书评

暂无评论

Implementation of the Barnard and Thompson algorithm on Transputers for SPOT stereo

Implementation of the Barnard and Thompson algorithm on Tran...

引用

iee colloquium on Motion and Stereopsis in Machine Vision

作者： K.A. Collins J.B.G. Roberts Royal Signals and Radar Establishment Malvern Worcestershire UK

The Transputer provides a powerful new capability in parallel computing which can readily be optimised for specific applications. The authors compare two alternative organisations of a large problem on a network of 21 Transputers, from the standpoints of speed, hardware efficiency and ease of programming. The authors programmed two highly parallel implementations of an algorithm which constitutes part of a 'real-time' system to generate terrain relief maps from satellite stereo image pairs. The particular algorithm used is the stereo matching strategy due to Barnard & Thompson (1980). The algorithm attempts to identify corresponding points in two image frames, in this case raw SPOT images, in order to refine initial estimates for the relative orientation parameters of the sensor in the two cases. The initial estimates are extracted from satellite header information, and the refined estimates are used to provide approximate epipolar image data for use with subsequent edge-based matching techniques. The authors describe one stage of the sequence implemented on the Transputers: apply the Barnard and Thompson algorithm to rough epipolar SPOT stereo pair, to obtain a spare disparity map.< >

关键词： Machine vision parallel architectures parallel programming

来源：评论

学校读者我要写书评

暂无评论

image DECORRELATION USING parallel ELEMENTS MADE FROM VLSI DEVICES.

IMAGE DECORRELATION USING PARALLEL ELEMENTS MADE FROM VLSI D...

引用

colloquium on VLSI for image processing.

作者： Bramley, R.G. Creasey, D.J. Univ of Birmingham Engl Univ of Birmingham Engl

Many operations required for image processing use transformations which are often derivatives of the discrete Fourier transform (DFT). The Fourier transform and its derivatives enable the two-dimensional transform to be separated into the double application of a one-dimensional transform and this simplifies the processing of images. This paper looks at a newly-produced VLSI device, the INMOS IMS A100, and investigates ways of reducing the memory input/output bandwidths and proposes an architecture which separates the control of data from the computational processes. The work is based upon the one-dimensional DFT because this has applications in many other areas of signal processing. Related transforms such as the discrete cosine transform and the Hartley transform can be mapped onto the same architecture. Since these related transforms are simpler than the DFT, they can result in a reduced hardware overhead.

关键词： INTEGRATED CIRCUITS, VLSI

来源：评论

学校读者我要写书评

暂无评论

image-RECONSTRUCTION USING THE TRANSPUTER

引用

iee PROCEEDINGS-I COMMUNICATIONS SPEECH AND VISION 1986年第3期133卷 99-104页

作者： KINGSWOOD, N DAGLESS, EL BELCHAMBER, RM BETTERIDGE, D LILLEY, T ROBERTS, JDM BP RES CTR SUNBURY THAMES TW16 7LNMIDDXENGLAND

image reconstruction by computer tomography provides a nonintrusive method of imaging the internal structure of objects. From measurements of radiation (e.g. X-rays or gamma rays) passed through an object, it is possible to reconstruct the internal structure. There is great interest in the potential of such methods in industrial applications but a number of problems need to be solved before these opportunities can be realised. The reconstruction process is computationally intensive and requires imaginative parallel processing algorithms to attain ‘real-time’performance. The work carried out has involved evaluating how these algorithms can be used in multiprocessor concurrent architectures to obtain rapid image reconstruction. A suitable computer architecture has been simulated in occam. This allows execution on a collection of transputers.

关键词： computer tomography parallel processing internal structure multiprocessor concurrent architectures computerised tomography transputer Digital signal processing microprocessor chips industrial applications Multiprocessing systems computerised picture processing image reconstruction nonintrusive method

来源：评论

学校读者我要写书评

暂无评论

image-RECONSTRUCTION USING THE TRANSPUTER

引用

iee PROCEEDINGS-E COMPUTERS AND DIGITAL TECHNIQUES 1986年第3期133卷 139-144页

作者： KINGSWOOD, N DAGLESS, EL BELCHAMBER, RM BETTERIDGE, D LILLEY, T ROBERTS, JDM BP RES CTR SUNBURY THAMES TW16 7LNMIDDXENGLAND

image reconstruction by computer tomography provides a nonintrusive method of imaging the internal structure of objects. From measurements of radiation (e.g. X-rays or gamma rays) passed through an object, it is possible to reconstruct the internal structure. There is great interest in the potential of such methods in industrial applications but a number of problems need to be solved before these opportunities can be realised. The reconstruction process is computationally intensive and requires imaginative parallel processing algorithms to attain ‘real-time’ performance. The work carried out has involved evaluating how these algorithms can be used in multiprocessor concurrent architectures to obtain rapid image reconstruction. A suitable computer architecture has been simulated in occam. This allows execution on a collection of transputers.

关键词： computer tomography internal structure imaging internal structure multiprocessor concurrent architectures computerised pattern recognition computer architecture computerised tomography Radiation and radioactivity applications in biomedicine Biology and medical computing transputer rapid image reconstruction industrial applications parallel processing algorithms Pattern recognition and computer vision equipment image reconstruction nonintrusive method

来源：评论

学校读者我要写书评

暂无评论

image RECONSTRUCTION USING parallel COMPUTATION.

IMAGE RECONSTRUCTION USING PARALLEL COMPUTATION.

引用

Second International Conference on image processing and Its applications.

作者： Kingswood, N. Univ of Bristol Engl Univ of Bristol Engl

ISBN: (纸本)0852963325

Computer tomography has been used in radiology and many other fields. Although several reconstruction algorithms have been used there are only two basic types: iterative and analytical. For industrial applications iterative methods are more suitable since they are more flexible in dealing with different transducer configurations and missing data. However iterative methods are considerably slower than analytical ones. This disadvantage may be overcome by using parallel algorithms running on parallel architectures. When using such algorithms the communications between the individual processes must be considered carefully if deadlock is to be prevented. In this paper a parallel architecture suitable for directly executing parallel iterative algorithms has been described. Since occam has been used ideas can be investigated and may then be executed on an array of transputers. This architecture is easily understood and gives a reasonable speed-up. However it uses a large number of transputers.

关键词： image processing

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：