A Biologically Inspired Approach for Fast Image Processing

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

Abstract

As features within an image may be present at many scales, application of feature detectors at multiple scales can improve accuracy of the detected localisation and orientation. As the scale and size of a feature detector increases, so does the computational complexity of implementation across the image domain. To address this issue we present a novel integral image for hexagonal pixel based images and associated multi-scale operator implementation that significantly speeds up the feature detection process. We demonstrate that this framework enables significantly faster computation than the use of conventional spiral convolution, the use of a neighbourhood address look-up table on hexagonal images.
LanguageEnglish
Title of host publicationUnknown Host Publication
Pages129-132
Number of pages4
Publication statusPublished - 20 May 2013
EventIAPR International Conference on Machine Vision Applications (MVA) - Kyoto, Japan
Duration: 20 May 2013 → …

Conference

ConferenceIAPR International Conference on Machine Vision Applications (MVA)
Period20/05/13 → …

Fingerprint

Image processing
Detectors
Convolution
Computational complexity
Pixels

Cite this

@inproceedings{9df7916915d8471ebcb51382e574015c,
title = "A Biologically Inspired Approach for Fast Image Processing",
abstract = "As features within an image may be present at many scales, application of feature detectors at multiple scales can improve accuracy of the detected localisation and orientation. As the scale and size of a feature detector increases, so does the computational complexity of implementation across the image domain. To address this issue we present a novel integral image for hexagonal pixel based images and associated multi-scale operator implementation that significantly speeds up the feature detection process. We demonstrate that this framework enables significantly faster computation than the use of conventional spiral convolution, the use of a neighbourhood address look-up table on hexagonal images.",
author = "SA Coleman and BW Scotney and B Gardiner",
year = "2013",
month = "5",
day = "20",
language = "English",
isbn = "978-4-901122-13-9",
pages = "129--132",
booktitle = "Unknown Host Publication",

}

Coleman, SA, Scotney, BW & Gardiner, B 2013, A Biologically Inspired Approach for Fast Image Processing. in Unknown Host Publication. pp. 129-132, IAPR International Conference on Machine Vision Applications (MVA), 20/05/13.

A Biologically Inspired Approach for Fast Image Processing. / Coleman, SA; Scotney, BW; Gardiner, B.

Unknown Host Publication. 2013. p. 129-132.

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

TY - GEN

T1 - A Biologically Inspired Approach for Fast Image Processing

AU - Coleman, SA

AU - Scotney, BW

AU - Gardiner, B

PY - 2013/5/20

Y1 - 2013/5/20

N2 - As features within an image may be present at many scales, application of feature detectors at multiple scales can improve accuracy of the detected localisation and orientation. As the scale and size of a feature detector increases, so does the computational complexity of implementation across the image domain. To address this issue we present a novel integral image for hexagonal pixel based images and associated multi-scale operator implementation that significantly speeds up the feature detection process. We demonstrate that this framework enables significantly faster computation than the use of conventional spiral convolution, the use of a neighbourhood address look-up table on hexagonal images.

AB - As features within an image may be present at many scales, application of feature detectors at multiple scales can improve accuracy of the detected localisation and orientation. As the scale and size of a feature detector increases, so does the computational complexity of implementation across the image domain. To address this issue we present a novel integral image for hexagonal pixel based images and associated multi-scale operator implementation that significantly speeds up the feature detection process. We demonstrate that this framework enables significantly faster computation than the use of conventional spiral convolution, the use of a neighbourhood address look-up table on hexagonal images.

M3 - Conference contribution

SN - 978-4-901122-13-9

SP - 129

EP - 132

BT - Unknown Host Publication

ER -