A reconfigurable and hierarchical parallel processing architecture: performance results for stereo vision

[1990] Proceedings. 10th International Conference on Pattern Recognition Pub Date : 1990-06-16 DOI:10.1109/ICPR.1990.119388

A. Choudhary, Subhodev Das, N. Ahuja, J. Patel

{"title":"A reconfigurable and hierarchical parallel processing architecture: performance results for stereo vision","authors":"A. Choudhary, Subhodev Das, N. Ahuja, J. Patel","doi":"10.1109/ICPR.1990.119388","DOIUrl":null,"url":null,"abstract":"A multiprocessor architecture called NETRA is discussed. It is highly reconfigurable and does not involve the use of complex interconnection schemes. The topology of this multiprocessor is recursively defined and is therefore easily scalable from small to large systems. It has a tree-type hierarchical architecture featuring leaf nodes that consist of a cluster of small but powerful processors connected via a programmable crossbar with selective broadcast capability. The architecture is simulated on a hypercube multiprocessor and the performance of one processor cluster is evaluated for stereo-vision tasks. The particular stereo algorithm selected for implementation requires computation of the two-dimensional fast Fourier transform (2-D FFT), template matching, histogram computation, and least-squares surface fitting. Static partitioning of data is used for the data-independent tasks such as 2-D FFT and dynamic scheduling, and load balancing is used for the data-dependent tasks of feature matching and disambiguation.<<ETX>>","PeriodicalId":135937,"journal":{"name":"[1990] Proceedings. 10th International Conference on Pattern Recognition","volume":"21 5 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1990-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"11","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"[1990] Proceedings. 10th International Conference on Pattern Recognition","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICPR.1990.119388","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 11

Abstract

A multiprocessor architecture called NETRA is discussed. It is highly reconfigurable and does not involve the use of complex interconnection schemes. The topology of this multiprocessor is recursively defined and is therefore easily scalable from small to large systems. It has a tree-type hierarchical architecture featuring leaf nodes that consist of a cluster of small but powerful processors connected via a programmable crossbar with selective broadcast capability. The architecture is simulated on a hypercube multiprocessor and the performance of one processor cluster is evaluated for stereo-vision tasks. The particular stereo algorithm selected for implementation requires computation of the two-dimensional fast Fourier transform (2-D FFT), template matching, histogram computation, and least-squares surface fitting. Static partitioning of data is used for the data-independent tasks such as 2-D FFT and dynamic scheduling, and load balancing is used for the data-dependent tasks of feature matching and disambiguation.<>

查看原文本刊更多论文

一种可重构的分层并行处理架构:立体视觉的性能结果

讨论了一种称为NETRA的多处理器体系结构。它是高度可重构的，不涉及使用复杂的互连方案。这个多处理器的拓扑是递归定义的，因此很容易从小型系统扩展到大型系统。它有一个树型层次结构，其特征是叶节点，由一组小而功能强大的处理器组成，这些处理器通过具有选择性广播功能的可编程横杆连接在一起。在一个超立方体多处理器上对该体系结构进行了仿真，并对一个处理器集群在立体视觉任务中的性能进行了评估。选择用于实现的特定立体算法需要计算二维快速傅里叶变换(2-D FFT)、模板匹配、直方图计算和最小二乘曲面拟合。二维FFT和动态调度等与数据无关的任务采用数据静态分区，特征匹配和消歧等与数据相关的任务采用负载均衡

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

[1990] Proceedings. 10th International Conference on Pattern Recognition

自引率

0.00%

发文量