A parallel algorithm for template matching on an SIMD mesh connected computer

Chaitali Chakrabarti, Joseph F. JaJa

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

2 Citations (Scopus)

Abstract

An efficient parallel algorithm to compute template matching of an N × N input image with an M × M template on a single-instruction multiple-data (SIMD) mesh-connected computer with P processors is proposed. The input image is mapped into the processor array such that each processor stores N 2/P data in the cyclic mode. The template values are circulated among the processors instead of being broadcast or stored in the processor memory. There is no movement of the intermediate results. The computation and the communication time complexity of the algorithm is O(M 2N 2/P) for all P in the range M 2 ≤ P ≤ N 2.

Original languageEnglish (US)
Title of host publicationProceedings - International Conference on Pattern Recognition
PublisherPubl by IEEE
Pages362-367
Number of pages6
Volume2
StatePublished - 1990
Externally publishedYes
EventProceedings of the 10th International Conference on Pattern Recognition - Atlantic City, NJ, USA
Duration: Jun 16 1990Jun 21 1990

Other

OtherProceedings of the 10th International Conference on Pattern Recognition
CityAtlantic City, NJ, USA
Period6/16/906/21/90

Fingerprint

Template matching
Parallel processing systems
Parallel algorithms
Data storage equipment
Communication

ASJC Scopus subject areas

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

Cite this

Chakrabarti, C., & JaJa, J. F. (1990). A parallel algorithm for template matching on an SIMD mesh connected computer. In Proceedings - International Conference on Pattern Recognition (Vol. 2, pp. 362-367). Publ by IEEE.

A parallel algorithm for template matching on an SIMD mesh connected computer. / Chakrabarti, Chaitali; JaJa, Joseph F.

Proceedings - International Conference on Pattern Recognition. Vol. 2 Publ by IEEE, 1990. p. 362-367.

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

Chakrabarti, C & JaJa, JF 1990, A parallel algorithm for template matching on an SIMD mesh connected computer. in Proceedings - International Conference on Pattern Recognition. vol. 2, Publ by IEEE, pp. 362-367, Proceedings of the 10th International Conference on Pattern Recognition, Atlantic City, NJ, USA, 6/16/90.
Chakrabarti C, JaJa JF. A parallel algorithm for template matching on an SIMD mesh connected computer. In Proceedings - International Conference on Pattern Recognition. Vol. 2. Publ by IEEE. 1990. p. 362-367
Chakrabarti, Chaitali ; JaJa, Joseph F. / A parallel algorithm for template matching on an SIMD mesh connected computer. Proceedings - International Conference on Pattern Recognition. Vol. 2 Publ by IEEE, 1990. pp. 362-367
@inproceedings{4efc6167a6b44d14bf11d9c491d99dc9,
title = "A parallel algorithm for template matching on an SIMD mesh connected computer",
abstract = "An efficient parallel algorithm to compute template matching of an N × N input image with an M × M template on a single-instruction multiple-data (SIMD) mesh-connected computer with P processors is proposed. The input image is mapped into the processor array such that each processor stores N 2/P data in the cyclic mode. The template values are circulated among the processors instead of being broadcast or stored in the processor memory. There is no movement of the intermediate results. The computation and the communication time complexity of the algorithm is O(M 2N 2/P) for all P in the range M 2 ≤ P ≤ N 2.",
author = "Chaitali Chakrabarti and JaJa, {Joseph F.}",
year = "1990",
language = "English (US)",
volume = "2",
pages = "362--367",
booktitle = "Proceedings - International Conference on Pattern Recognition",
publisher = "Publ by IEEE",

}

TY - GEN

T1 - A parallel algorithm for template matching on an SIMD mesh connected computer

AU - Chakrabarti, Chaitali

AU - JaJa, Joseph F.

PY - 1990

Y1 - 1990

N2 - An efficient parallel algorithm to compute template matching of an N × N input image with an M × M template on a single-instruction multiple-data (SIMD) mesh-connected computer with P processors is proposed. The input image is mapped into the processor array such that each processor stores N 2/P data in the cyclic mode. The template values are circulated among the processors instead of being broadcast or stored in the processor memory. There is no movement of the intermediate results. The computation and the communication time complexity of the algorithm is O(M 2N 2/P) for all P in the range M 2 ≤ P ≤ N 2.

AB - An efficient parallel algorithm to compute template matching of an N × N input image with an M × M template on a single-instruction multiple-data (SIMD) mesh-connected computer with P processors is proposed. The input image is mapped into the processor array such that each processor stores N 2/P data in the cyclic mode. The template values are circulated among the processors instead of being broadcast or stored in the processor memory. There is no movement of the intermediate results. The computation and the communication time complexity of the algorithm is O(M 2N 2/P) for all P in the range M 2 ≤ P ≤ N 2.

UR - http://www.scopus.com/inward/record.url?scp=0025560639&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0025560639&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0025560639

VL - 2

SP - 362

EP - 367

BT - Proceedings - International Conference on Pattern Recognition

PB - Publ by IEEE

ER -