Multiresolution gabor feature extraction for real time applications

Yong Cheol Peter Cho, Nandhini Chadramoorthy, Kevin M. Irick, Vijayknshnan Narayanan

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

3 Citations (Scopus)

Abstract

Multiresolution Gabor filters are used for feature extraction for a variety of applications. Most hardware implementations have focused on iterative mechanisms on fixed hardware for implementing the different levels of resolution. In contrast, we present a configurable architecture that enhances the resource utilization of the hardware fabric. Our results show that our implementation achieves real-time performance on 2048×1536 images and exhibits 6 times speed up over a GPU implementation. Further, our FPGA implementation achieves an energy-efficiency of processing 0.4 fps/W as compared to the GPU that achieves 0.036 fps/W.

Original languageEnglish (US)
Title of host publicationProceedings - 2012 IEEE Workshop on Signal Processing Systems, SiPS 2012
Pages55-60
Number of pages6
DOIs
StatePublished - Dec 1 2012
Event2012 IEEE Workshop on Signal Processing Systems, SiPS 2012 - Quebec City, QC, Canada
Duration: Oct 17 2012Oct 19 2012

Publication series

NameIEEE Workshop on Signal Processing Systems, SiPS: Design and Implementation
ISSN (Print)1520-6130

Other

Other2012 IEEE Workshop on Signal Processing Systems, SiPS 2012
CountryCanada
CityQuebec City, QC
Period10/17/1210/19/12

Fingerprint

Multiresolution
Feature Extraction
Feature extraction
Hardware
FPGA Implementation
Gabor Filter
Hardware Implementation
Energy Efficiency
Gabor filters
Speedup
Real-time
Resources
Energy efficiency
Field programmable gate arrays (FPGA)
Processing
Graphics processing unit
Architecture

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering
  • Signal Processing
  • Applied Mathematics
  • Hardware and Architecture

Cite this

Cho, Y. C. P., Chadramoorthy, N., Irick, K. M., & Narayanan, V. (2012). Multiresolution gabor feature extraction for real time applications. In Proceedings - 2012 IEEE Workshop on Signal Processing Systems, SiPS 2012 (pp. 55-60). [6363183] (IEEE Workshop on Signal Processing Systems, SiPS: Design and Implementation). https://doi.org/10.1109/SiPS.2012.56
Cho, Yong Cheol Peter ; Chadramoorthy, Nandhini ; Irick, Kevin M. ; Narayanan, Vijayknshnan. / Multiresolution gabor feature extraction for real time applications. Proceedings - 2012 IEEE Workshop on Signal Processing Systems, SiPS 2012. 2012. pp. 55-60 (IEEE Workshop on Signal Processing Systems, SiPS: Design and Implementation).
@inproceedings{09afa5e908b24bd497d1da0c8fa9672c,
title = "Multiresolution gabor feature extraction for real time applications",
abstract = "Multiresolution Gabor filters are used for feature extraction for a variety of applications. Most hardware implementations have focused on iterative mechanisms on fixed hardware for implementing the different levels of resolution. In contrast, we present a configurable architecture that enhances the resource utilization of the hardware fabric. Our results show that our implementation achieves real-time performance on 2048×1536 images and exhibits 6 times speed up over a GPU implementation. Further, our FPGA implementation achieves an energy-efficiency of processing 0.4 fps/W as compared to the GPU that achieves 0.036 fps/W.",
author = "Cho, {Yong Cheol Peter} and Nandhini Chadramoorthy and Irick, {Kevin M.} and Vijayknshnan Narayanan",
year = "2012",
month = "12",
day = "1",
doi = "10.1109/SiPS.2012.56",
language = "English (US)",
isbn = "9780769548562",
series = "IEEE Workshop on Signal Processing Systems, SiPS: Design and Implementation",
pages = "55--60",
booktitle = "Proceedings - 2012 IEEE Workshop on Signal Processing Systems, SiPS 2012",

}

Cho, YCP, Chadramoorthy, N, Irick, KM & Narayanan, V 2012, Multiresolution gabor feature extraction for real time applications. in Proceedings - 2012 IEEE Workshop on Signal Processing Systems, SiPS 2012., 6363183, IEEE Workshop on Signal Processing Systems, SiPS: Design and Implementation, pp. 55-60, 2012 IEEE Workshop on Signal Processing Systems, SiPS 2012, Quebec City, QC, Canada, 10/17/12. https://doi.org/10.1109/SiPS.2012.56

Multiresolution gabor feature extraction for real time applications. / Cho, Yong Cheol Peter; Chadramoorthy, Nandhini; Irick, Kevin M.; Narayanan, Vijayknshnan.

Proceedings - 2012 IEEE Workshop on Signal Processing Systems, SiPS 2012. 2012. p. 55-60 6363183 (IEEE Workshop on Signal Processing Systems, SiPS: Design and Implementation).

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

TY - GEN

T1 - Multiresolution gabor feature extraction for real time applications

AU - Cho, Yong Cheol Peter

AU - Chadramoorthy, Nandhini

AU - Irick, Kevin M.

AU - Narayanan, Vijayknshnan

PY - 2012/12/1

Y1 - 2012/12/1

N2 - Multiresolution Gabor filters are used for feature extraction for a variety of applications. Most hardware implementations have focused on iterative mechanisms on fixed hardware for implementing the different levels of resolution. In contrast, we present a configurable architecture that enhances the resource utilization of the hardware fabric. Our results show that our implementation achieves real-time performance on 2048×1536 images and exhibits 6 times speed up over a GPU implementation. Further, our FPGA implementation achieves an energy-efficiency of processing 0.4 fps/W as compared to the GPU that achieves 0.036 fps/W.

AB - Multiresolution Gabor filters are used for feature extraction for a variety of applications. Most hardware implementations have focused on iterative mechanisms on fixed hardware for implementing the different levels of resolution. In contrast, we present a configurable architecture that enhances the resource utilization of the hardware fabric. Our results show that our implementation achieves real-time performance on 2048×1536 images and exhibits 6 times speed up over a GPU implementation. Further, our FPGA implementation achieves an energy-efficiency of processing 0.4 fps/W as compared to the GPU that achieves 0.036 fps/W.

UR - http://www.scopus.com/inward/record.url?scp=84875291058&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84875291058&partnerID=8YFLogxK

U2 - 10.1109/SiPS.2012.56

DO - 10.1109/SiPS.2012.56

M3 - Conference contribution

AN - SCOPUS:84875291058

SN - 9780769548562

T3 - IEEE Workshop on Signal Processing Systems, SiPS: Design and Implementation

SP - 55

EP - 60

BT - Proceedings - 2012 IEEE Workshop on Signal Processing Systems, SiPS 2012

ER -

Cho YCP, Chadramoorthy N, Irick KM, Narayanan V. Multiresolution gabor feature extraction for real time applications. In Proceedings - 2012 IEEE Workshop on Signal Processing Systems, SiPS 2012. 2012. p. 55-60. 6363183. (IEEE Workshop on Signal Processing Systems, SiPS: Design and Implementation). https://doi.org/10.1109/SiPS.2012.56