Accelerating image processing algorithms based on the reuse of spatial patterns

Farzad Khalvati; Mark D. Aagaard; Hamid R. Tizhoosh

doi:10.1109/CCECE.2007.50

Accelerating image processing algorithms based on the reuse of spatial patterns

Farzad Khalvati, Mark D. Aagaard, Hamid R. Tizhoosh

Artificial Intelligence and Informatics

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

This paper presents window memoization, a performance optimization technique for convolution-based image processing algorithms. Window memoization exploits the repetitive nature of image data to reduce the number of calculations required for image processing algorithms and hence, it improves the performance. We applied window memoization to a chain of image processing algorithms that includes median filter, Kirsch edge detector and local edge filling. We found that a large portion of the calculations performed on pixel neighborhoods can be skipped and instead, previously calculated results can be reused. The typical speedups were in the range of 1.6x to 2.8x.

Original language	English (US)
Title of host publication	2007 Canadian Conference on Electrical and Computer Engineering, CCECD
Pages	172-175
Number of pages	4
DOIs	https://doi.org/10.1109/CCECE.2007.50
State	Published - 2007
Event	2007 Canadian Conference on Electrical and Computer Engineering, CCECD - Vancouver, BC, Canada Duration: Apr 22 2007 → Apr 26 2007

Publication series

Name	Canadian Conference on Electrical and Computer Engineering
ISSN (Print)	0840-7789

Other

Other	2007 Canadian Conference on Electrical and Computer Engineering, CCECD
Country/Territory	Canada
City	Vancouver, BC
Period	4/22/07 → 4/26/07

ASJC Scopus subject areas

Hardware and Architecture
Electrical and Electronic Engineering

Access to Document

10.1109/CCECE.2007.50

Cite this

Accelerating image processing algorithms based on the reuse of spatial patterns. / Khalvati, Farzad; Aagaard, Mark D.; Tizhoosh, Hamid R.
2007 Canadian Conference on Electrical and Computer Engineering, CCECD. 2007. p. 172-175 4232709 (Canadian Conference on Electrical and Computer Engineering).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Khalvati, F, Aagaard, MD & Tizhoosh, HR 2007, Accelerating image processing algorithms based on the reuse of spatial patterns. in 2007 Canadian Conference on Electrical and Computer Engineering, CCECD., 4232709, Canadian Conference on Electrical and Computer Engineering, pp. 172-175, 2007 Canadian Conference on Electrical and Computer Engineering, CCECD, Vancouver, BC, Canada, 4/22/07. https://doi.org/10.1109/CCECE.2007.50

@inproceedings{23833f398e874ff1be4e6b53403d2d92,

title = "Accelerating image processing algorithms based on the reuse of spatial patterns",

abstract = "This paper presents window memoization, a performance optimization technique for convolution-based image processing algorithms. Window memoization exploits the repetitive nature of image data to reduce the number of calculations required for image processing algorithms and hence, it improves the performance. We applied window memoization to a chain of image processing algorithms that includes median filter, Kirsch edge detector and local edge filling. We found that a large portion of the calculations performed on pixel neighborhoods can be skipped and instead, previously calculated results can be reused. The typical speedups were in the range of 1.6x to 2.8x.",

author = "Farzad Khalvati and Aagaard, {Mark D.} and Tizhoosh, {Hamid R.}",

year = "2007",

doi = "10.1109/CCECE.2007.50",

language = "English (US)",

isbn = "1424410215",

series = "Canadian Conference on Electrical and Computer Engineering",

pages = "172--175",

booktitle = "2007 Canadian Conference on Electrical and Computer Engineering, CCECD",

note = "2007 Canadian Conference on Electrical and Computer Engineering, CCECD ; Conference date: 22-04-2007 Through 26-04-2007",

}

TY - GEN

T1 - Accelerating image processing algorithms based on the reuse of spatial patterns

AU - Khalvati, Farzad

AU - Aagaard, Mark D.

AU - Tizhoosh, Hamid R.

PY - 2007

Y1 - 2007

N2 - This paper presents window memoization, a performance optimization technique for convolution-based image processing algorithms. Window memoization exploits the repetitive nature of image data to reduce the number of calculations required for image processing algorithms and hence, it improves the performance. We applied window memoization to a chain of image processing algorithms that includes median filter, Kirsch edge detector and local edge filling. We found that a large portion of the calculations performed on pixel neighborhoods can be skipped and instead, previously calculated results can be reused. The typical speedups were in the range of 1.6x to 2.8x.

AB - This paper presents window memoization, a performance optimization technique for convolution-based image processing algorithms. Window memoization exploits the repetitive nature of image data to reduce the number of calculations required for image processing algorithms and hence, it improves the performance. We applied window memoization to a chain of image processing algorithms that includes median filter, Kirsch edge detector and local edge filling. We found that a large portion of the calculations performed on pixel neighborhoods can be skipped and instead, previously calculated results can be reused. The typical speedups were in the range of 1.6x to 2.8x.

UR - http://www.scopus.com/inward/record.url?scp=48749132747&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=48749132747&partnerID=8YFLogxK

U2 - 10.1109/CCECE.2007.50

DO - 10.1109/CCECE.2007.50

M3 - Conference contribution

AN - SCOPUS:48749132747

SN - 1424410215

SN - 9781424410217

T3 - Canadian Conference on Electrical and Computer Engineering

SP - 172

EP - 175

BT - 2007 Canadian Conference on Electrical and Computer Engineering, CCECD

T2 - 2007 Canadian Conference on Electrical and Computer Engineering, CCECD

Y2 - 22 April 2007 through 26 April 2007

ER -

Accelerating image processing algorithms based on the reuse of spatial patterns

Abstract

Publication series

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this