Orthogonal access architectures and reduced meshes for parallel image computations

Hussein Alnuweiri, V. K.Prasanna Kumar

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A class of orthogonal-access parallel organizations is studied for applications in image and vision analysis. These architectures consist of a massive memory and a reduced number of processors which access the shared memory. The memory can be envisaged as an array of memory modules in the k-dimensional space, with each row of modules along a certain dimension connected to one bus. Each processor has access to one bus along each dimension. It is shown that these organizations are communication-efficient and can provide processor-time optimal solutions to a wide class of image and vision problems. In the two-dimensional case, the basic organization has n processors and an n × n memory array which can hold an n × n image, and it provides O(n) time solution to several image computations including: histograming, histogram equalization, computing connected components, convexity problems, and computing distances. Such problems also take O(n) time on a two-dimensional mesh with n2 processors. For the general k-dimensional case, a class of orthogonal data movement operations can be implemented on such organizations to yield processor-time optimal image and vision algorithms.

Original languageEnglish
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
EditorsJoydeep Ghosh, G.Colin Harrison
PublisherPubl by Int Soc for Optical Engineering
Pages212-223
Number of pages12
Volume1246
ISBN (Print)0819402931
Publication statusPublished - 1 Dec 1990
Externally publishedYes
EventParallel Architectures for Image Processing - Santa Clara, CA, USA
Duration: 14 Feb 199015 Feb 1990

Other

OtherParallel Architectures for Image Processing
CitySanta Clara, CA, USA
Period14/2/9015/2/90

Fingerprint

central processing units
mesh
Mesh
Data storage equipment
Histogram Equalization
modules
Module
Computing
convexity
Shared Memory
Connected Components
Convexity
image analysis
histograms
Optimal Solution
Architecture
communication
Communication
Class
Vision

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering

Cite this

Alnuweiri, H., & Kumar, V. K. P. (1990). Orthogonal access architectures and reduced meshes for parallel image computations. In J. Ghosh, & G. C. Harrison (Eds.), Proceedings of SPIE - The International Society for Optical Engineering (Vol. 1246, pp. 212-223). Publ by Int Soc for Optical Engineering.

Orthogonal access architectures and reduced meshes for parallel image computations. / Alnuweiri, Hussein; Kumar, V. K.Prasanna.

Proceedings of SPIE - The International Society for Optical Engineering. ed. / Joydeep Ghosh; G.Colin Harrison. Vol. 1246 Publ by Int Soc for Optical Engineering, 1990. p. 212-223.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Alnuweiri, H & Kumar, VKP 1990, Orthogonal access architectures and reduced meshes for parallel image computations. in J Ghosh & GC Harrison (eds), Proceedings of SPIE - The International Society for Optical Engineering. vol. 1246, Publ by Int Soc for Optical Engineering, pp. 212-223, Parallel Architectures for Image Processing, Santa Clara, CA, USA, 14/2/90.
Alnuweiri H, Kumar VKP. Orthogonal access architectures and reduced meshes for parallel image computations. In Ghosh J, Harrison GC, editors, Proceedings of SPIE - The International Society for Optical Engineering. Vol. 1246. Publ by Int Soc for Optical Engineering. 1990. p. 212-223
Alnuweiri, Hussein ; Kumar, V. K.Prasanna. / Orthogonal access architectures and reduced meshes for parallel image computations. Proceedings of SPIE - The International Society for Optical Engineering. editor / Joydeep Ghosh ; G.Colin Harrison. Vol. 1246 Publ by Int Soc for Optical Engineering, 1990. pp. 212-223
@inproceedings{075a7a3318c84dd9baf42d1a097318bf,
title = "Orthogonal access architectures and reduced meshes for parallel image computations",
abstract = "A class of orthogonal-access parallel organizations is studied for applications in image and vision analysis. These architectures consist of a massive memory and a reduced number of processors which access the shared memory. The memory can be envisaged as an array of memory modules in the k-dimensional space, with each row of modules along a certain dimension connected to one bus. Each processor has access to one bus along each dimension. It is shown that these organizations are communication-efficient and can provide processor-time optimal solutions to a wide class of image and vision problems. In the two-dimensional case, the basic organization has n processors and an n × n memory array which can hold an n × n image, and it provides O(n) time solution to several image computations including: histograming, histogram equalization, computing connected components, convexity problems, and computing distances. Such problems also take O(n) time on a two-dimensional mesh with n2 processors. For the general k-dimensional case, a class of orthogonal data movement operations can be implemented on such organizations to yield processor-time optimal image and vision algorithms.",
author = "Hussein Alnuweiri and Kumar, {V. K.Prasanna}",
year = "1990",
month = "12",
day = "1",
language = "English",
isbn = "0819402931",
volume = "1246",
pages = "212--223",
editor = "Joydeep Ghosh and G.Colin Harrison",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",
publisher = "Publ by Int Soc for Optical Engineering",

}

TY - GEN

T1 - Orthogonal access architectures and reduced meshes for parallel image computations

AU - Alnuweiri, Hussein

AU - Kumar, V. K.Prasanna

PY - 1990/12/1

Y1 - 1990/12/1

N2 - A class of orthogonal-access parallel organizations is studied for applications in image and vision analysis. These architectures consist of a massive memory and a reduced number of processors which access the shared memory. The memory can be envisaged as an array of memory modules in the k-dimensional space, with each row of modules along a certain dimension connected to one bus. Each processor has access to one bus along each dimension. It is shown that these organizations are communication-efficient and can provide processor-time optimal solutions to a wide class of image and vision problems. In the two-dimensional case, the basic organization has n processors and an n × n memory array which can hold an n × n image, and it provides O(n) time solution to several image computations including: histograming, histogram equalization, computing connected components, convexity problems, and computing distances. Such problems also take O(n) time on a two-dimensional mesh with n2 processors. For the general k-dimensional case, a class of orthogonal data movement operations can be implemented on such organizations to yield processor-time optimal image and vision algorithms.

AB - A class of orthogonal-access parallel organizations is studied for applications in image and vision analysis. These architectures consist of a massive memory and a reduced number of processors which access the shared memory. The memory can be envisaged as an array of memory modules in the k-dimensional space, with each row of modules along a certain dimension connected to one bus. Each processor has access to one bus along each dimension. It is shown that these organizations are communication-efficient and can provide processor-time optimal solutions to a wide class of image and vision problems. In the two-dimensional case, the basic organization has n processors and an n × n memory array which can hold an n × n image, and it provides O(n) time solution to several image computations including: histograming, histogram equalization, computing connected components, convexity problems, and computing distances. Such problems also take O(n) time on a two-dimensional mesh with n2 processors. For the general k-dimensional case, a class of orthogonal data movement operations can be implemented on such organizations to yield processor-time optimal image and vision algorithms.

UR - http://www.scopus.com/inward/record.url?scp=0025544373&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0025544373&partnerID=8YFLogxK

M3 - Conference contribution

SN - 0819402931

VL - 1246

SP - 212

EP - 223

BT - Proceedings of SPIE - The International Society for Optical Engineering

A2 - Ghosh, Joydeep

A2 - Harrison, G.Colin

PB - Publ by Int Soc for Optical Engineering

ER -