A colour hit-or-miss transform based on a rank ordered distance measure

Fraser Macfarlane, Paul Murray, Stephen Marshall, Benjamin Perret, Adrian Evans, Henry White

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

Abstract

The Hit-or-Miss Transform (HMT) is a powerful morphological operation that can be utilised in many digital image analysis problems. Its original binary definition and its extension to grey-level images have seen it applied to various template matching and object detection tasks. However, further extending the transform to incorporate colour or multivariate images is problematic since there is no general or intuitive way of ordering data which allows the formal definition of morphological operations in the traditional manner. In this paper, instead of following the usual strategy for Mathematical Morphology, based on the definition of a total order in the colour space, we propose a transform that relies on a colour or multivariate distance measure. As with the traditional HMT operator, our proposed transform uses two structuring elements (SE) - one for the foreground and one for the background - and retains the idea that a good fitting is obtained when the foreground SE is a close match to the image and the background SE matches the image complement. This allows for both flat and non-flat structuring elements to be used in object detection. Furthermore, the use of ranking operations on the computed distances allows the operator to be robust to noise and partial occlusion of objects.
LanguageEnglish
Title of host publicationProceedings of 26th European Signal Processing Conference, EUSIPCO 2018
PublisherIEEE
Pages588-592
Number of pages5
Volume2018-September
ISBN (Electronic)9789082797015
DOIs
StatusPublished - 29 Nov 2018

Fingerprint

Color
Mathematical transformations
Mathematical morphology
Template matching
Image analysis
Mathematical operators
Object detection

Keywords

  • Hit-or-Miss Transform
  • Image processing
  • Mathematical morphology
  • Object detection
  • Template matching

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Macfarlane, F., Murray, P., Marshall, S., Perret, B., Evans, A., & White, H. (2018). A colour hit-or-miss transform based on a rank ordered distance measure. In Proceedings of 26th European Signal Processing Conference, EUSIPCO 2018 (Vol. 2018-September, pp. 588-592). [8553050] IEEE. https://doi.org/10.23919/EUSIPCO.2018.8553050

A colour hit-or-miss transform based on a rank ordered distance measure. / Macfarlane, Fraser; Murray, Paul; Marshall, Stephen; Perret, Benjamin; Evans, Adrian; White, Henry.

Proceedings of 26th European Signal Processing Conference, EUSIPCO 2018. Vol. 2018-September IEEE, 2018. p. 588-592 8553050.

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

Macfarlane, F, Murray, P, Marshall, S, Perret, B, Evans, A & White, H 2018, A colour hit-or-miss transform based on a rank ordered distance measure. in Proceedings of 26th European Signal Processing Conference, EUSIPCO 2018. vol. 2018-September, 8553050, IEEE, pp. 588-592. https://doi.org/10.23919/EUSIPCO.2018.8553050
Macfarlane F, Murray P, Marshall S, Perret B, Evans A, White H. A colour hit-or-miss transform based on a rank ordered distance measure. In Proceedings of 26th European Signal Processing Conference, EUSIPCO 2018. Vol. 2018-September. IEEE. 2018. p. 588-592. 8553050 https://doi.org/10.23919/EUSIPCO.2018.8553050
Macfarlane, Fraser ; Murray, Paul ; Marshall, Stephen ; Perret, Benjamin ; Evans, Adrian ; White, Henry. / A colour hit-or-miss transform based on a rank ordered distance measure. Proceedings of 26th European Signal Processing Conference, EUSIPCO 2018. Vol. 2018-September IEEE, 2018. pp. 588-592
@inproceedings{e2d4753f1c444e8881568868e4b24e7a,
title = "A colour hit-or-miss transform based on a rank ordered distance measure",
abstract = "The Hit-or-Miss Transform (HMT) is a powerful morphological operation that can be utilised in many digital image analysis problems. Its original binary definition and its extension to grey-level images have seen it applied to various template matching and object detection tasks. However, further extending the transform to incorporate colour or multivariate images is problematic since there is no general or intuitive way of ordering data which allows the formal definition of morphological operations in the traditional manner. In this paper, instead of following the usual strategy for Mathematical Morphology, based on the definition of a total order in the colour space, we propose a transform that relies on a colour or multivariate distance measure. As with the traditional HMT operator, our proposed transform uses two structuring elements (SE) - one for the foreground and one for the background - and retains the idea that a good fitting is obtained when the foreground SE is a close match to the image and the background SE matches the image complement. This allows for both flat and non-flat structuring elements to be used in object detection. Furthermore, the use of ranking operations on the computed distances allows the operator to be robust to noise and partial occlusion of objects.",
keywords = "Hit-or-Miss Transform, Image processing, Mathematical morphology, Object detection, Template matching",
author = "Fraser Macfarlane and Paul Murray and Stephen Marshall and Benjamin Perret and Adrian Evans and Henry White",
year = "2018",
month = "11",
day = "29",
doi = "10.23919/EUSIPCO.2018.8553050",
language = "English",
volume = "2018-September",
pages = "588--592",
booktitle = "Proceedings of 26th European Signal Processing Conference, EUSIPCO 2018",
publisher = "IEEE",
address = "USA United States",

}

TY - GEN

T1 - A colour hit-or-miss transform based on a rank ordered distance measure

AU - Macfarlane, Fraser

AU - Murray, Paul

AU - Marshall, Stephen

AU - Perret, Benjamin

AU - Evans, Adrian

AU - White, Henry

PY - 2018/11/29

Y1 - 2018/11/29

N2 - The Hit-or-Miss Transform (HMT) is a powerful morphological operation that can be utilised in many digital image analysis problems. Its original binary definition and its extension to grey-level images have seen it applied to various template matching and object detection tasks. However, further extending the transform to incorporate colour or multivariate images is problematic since there is no general or intuitive way of ordering data which allows the formal definition of morphological operations in the traditional manner. In this paper, instead of following the usual strategy for Mathematical Morphology, based on the definition of a total order in the colour space, we propose a transform that relies on a colour or multivariate distance measure. As with the traditional HMT operator, our proposed transform uses two structuring elements (SE) - one for the foreground and one for the background - and retains the idea that a good fitting is obtained when the foreground SE is a close match to the image and the background SE matches the image complement. This allows for both flat and non-flat structuring elements to be used in object detection. Furthermore, the use of ranking operations on the computed distances allows the operator to be robust to noise and partial occlusion of objects.

AB - The Hit-or-Miss Transform (HMT) is a powerful morphological operation that can be utilised in many digital image analysis problems. Its original binary definition and its extension to grey-level images have seen it applied to various template matching and object detection tasks. However, further extending the transform to incorporate colour or multivariate images is problematic since there is no general or intuitive way of ordering data which allows the formal definition of morphological operations in the traditional manner. In this paper, instead of following the usual strategy for Mathematical Morphology, based on the definition of a total order in the colour space, we propose a transform that relies on a colour or multivariate distance measure. As with the traditional HMT operator, our proposed transform uses two structuring elements (SE) - one for the foreground and one for the background - and retains the idea that a good fitting is obtained when the foreground SE is a close match to the image and the background SE matches the image complement. This allows for both flat and non-flat structuring elements to be used in object detection. Furthermore, the use of ranking operations on the computed distances allows the operator to be robust to noise and partial occlusion of objects.

KW - Hit-or-Miss Transform

KW - Image processing

KW - Mathematical morphology

KW - Object detection

KW - Template matching

UR - http://www.scopus.com/inward/record.url?scp=85059815074&partnerID=8YFLogxK

U2 - 10.23919/EUSIPCO.2018.8553050

DO - 10.23919/EUSIPCO.2018.8553050

M3 - Conference contribution

VL - 2018-September

SP - 588

EP - 592

BT - Proceedings of 26th European Signal Processing Conference, EUSIPCO 2018

PB - IEEE

ER -