Optimal image algorithms on an orthogonally-connected memory-based architecture

Hussein Alnuweiri, V. K Prasanna Kumar

Research output: Contribution to journalConference article

1 Citation (Scopus)

Abstract

Processor-time optimal algorithms are presented for several image and vision problems. A parallel architecture which combines an orthogonally accessed memory with a linear array structure is used. The organization has p processors and a memory of size O(n2) locations. The number of processors p can vary over the range [1, n3/2] while providing optimal speedup for several problems in image analysis and vision. Such problems include labeling connected regions, computing minimum convex containers of regions, and computing nearest neighbors of pixels and regions. Optimal algorithms are presented for histogramming and computing the Hough transform. Such problems arise in medium-level vision and require global operations or dense data movement. It is shown that for these types of problems, the proposed organization is superior to the mesh and pyramid organizations.

Original languageEnglish
Pages (from-to)350-355
Number of pages6
JournalProceedings - International Conference on Pattern Recognition
Volume2
Publication statusPublished - 1 Dec 1990
Externally publishedYes

Fingerprint

Data storage equipment
Hough transforms
Parallel architectures
Labeling
Image analysis
Containers
Pixels

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Hardware and Architecture
  • Electrical and Electronic Engineering

Cite this

Optimal image algorithms on an orthogonally-connected memory-based architecture. / Alnuweiri, Hussein; Kumar, V. K Prasanna.

In: Proceedings - International Conference on Pattern Recognition, Vol. 2, 01.12.1990, p. 350-355.

Research output: Contribution to journalConference article

@article{6335b1cc86704ea0a3c0d8f2fbefbb33,
title = "Optimal image algorithms on an orthogonally-connected memory-based architecture",
abstract = "Processor-time optimal algorithms are presented for several image and vision problems. A parallel architecture which combines an orthogonally accessed memory with a linear array structure is used. The organization has p processors and a memory of size O(n2) locations. The number of processors p can vary over the range [1, n3/2] while providing optimal speedup for several problems in image analysis and vision. Such problems include labeling connected regions, computing minimum convex containers of regions, and computing nearest neighbors of pixels and regions. Optimal algorithms are presented for histogramming and computing the Hough transform. Such problems arise in medium-level vision and require global operations or dense data movement. It is shown that for these types of problems, the proposed organization is superior to the mesh and pyramid organizations.",
author = "Hussein Alnuweiri and Kumar, {V. K Prasanna}",
year = "1990",
month = "12",
day = "1",
language = "English",
volume = "2",
pages = "350--355",
journal = "Proceedings - International Conference on Pattern Recognition",
issn = "1051-4651",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Optimal image algorithms on an orthogonally-connected memory-based architecture

AU - Alnuweiri, Hussein

AU - Kumar, V. K Prasanna

PY - 1990/12/1

Y1 - 1990/12/1

N2 - Processor-time optimal algorithms are presented for several image and vision problems. A parallel architecture which combines an orthogonally accessed memory with a linear array structure is used. The organization has p processors and a memory of size O(n2) locations. The number of processors p can vary over the range [1, n3/2] while providing optimal speedup for several problems in image analysis and vision. Such problems include labeling connected regions, computing minimum convex containers of regions, and computing nearest neighbors of pixels and regions. Optimal algorithms are presented for histogramming and computing the Hough transform. Such problems arise in medium-level vision and require global operations or dense data movement. It is shown that for these types of problems, the proposed organization is superior to the mesh and pyramid organizations.

AB - Processor-time optimal algorithms are presented for several image and vision problems. A parallel architecture which combines an orthogonally accessed memory with a linear array structure is used. The organization has p processors and a memory of size O(n2) locations. The number of processors p can vary over the range [1, n3/2] while providing optimal speedup for several problems in image analysis and vision. Such problems include labeling connected regions, computing minimum convex containers of regions, and computing nearest neighbors of pixels and regions. Optimal algorithms are presented for histogramming and computing the Hough transform. Such problems arise in medium-level vision and require global operations or dense data movement. It is shown that for these types of problems, the proposed organization is superior to the mesh and pyramid organizations.

UR - http://www.scopus.com/inward/record.url?scp=0025543663&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0025543663&partnerID=8YFLogxK

M3 - Conference article

VL - 2

SP - 350

EP - 355

JO - Proceedings - International Conference on Pattern Recognition

JF - Proceedings - International Conference on Pattern Recognition

SN - 1051-4651

ER -