Segmentation algorithm for objects with very low edge contrast

Stephen McCarthy, Timothy C. Miller, Isaac N. Bankman

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

Abstract

We present an algorithm for segmentation of objects with very low edge contrast, such as microcalcifications in mammogram images. Most methods used to segment microcalcifications have algorithmic aspects that could raise operational difficulties, such as thresholds or windows that must be selected manually, or parametric models of the data. The presented algorithm does not use any of these techniques and does not require that any parameters be set by a user. It builds upon an earlier algorithm presented in, but is much faster and also applicable to a wider range of objects to be segmented. The algorithm's approach is based on the extension of radial intensity profiles from a given seed point to the edge of the image. A first derivative analysis is used to find an edge point pixel along each directional intensity profile. These points are connected and the resulting object border is filled using a constrained dilatation operation to form a complete region. Results from the tested mammography images indicate that the segmented regions compare closely to those expected from visual inspection.

Original languageEnglish (US)
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
EditorsD.P. Casasent, E.L. Hall, J. Roening
Pages162-168
Number of pages7
Volume5608
DOIs
StatePublished - 2004
EventIntellignt Robots and Computer Vision XXII: Algorithms, Techniques, and Active Vision - Philadelphia, PA, United States
Duration: Oct 25 2004Oct 27 2004

Other

OtherIntellignt Robots and Computer Vision XXII: Algorithms, Techniques, and Active Vision
CountryUnited States
CityPhiladelphia, PA
Period10/25/0410/27/04

Fingerprint

Mammography
profiles
borders
Seed
seeds
inspection
Inspection
Pixels
pixels
Derivatives
thresholds

Keywords

  • Automated image analysis system
  • Fast algorithm
  • Image segmentation
  • Low contrast image

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

McCarthy, S., Miller, T. C., & Bankman, I. N. (2004). Segmentation algorithm for objects with very low edge contrast. In D. P. Casasent, E. L. Hall, & J. Roening (Eds.), Proceedings of SPIE - The International Society for Optical Engineering (Vol. 5608, pp. 162-168). [19] https://doi.org/10.1117/12.580136

Segmentation algorithm for objects with very low edge contrast. / McCarthy, Stephen; Miller, Timothy C.; Bankman, Isaac N.

Proceedings of SPIE - The International Society for Optical Engineering. ed. / D.P. Casasent; E.L. Hall; J. Roening. Vol. 5608 2004. p. 162-168 19.

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

McCarthy, S, Miller, TC & Bankman, IN 2004, Segmentation algorithm for objects with very low edge contrast. in DP Casasent, EL Hall & J Roening (eds), Proceedings of SPIE - The International Society for Optical Engineering. vol. 5608, 19, pp. 162-168, Intellignt Robots and Computer Vision XXII: Algorithms, Techniques, and Active Vision, Philadelphia, PA, United States, 10/25/04. https://doi.org/10.1117/12.580136
McCarthy S, Miller TC, Bankman IN. Segmentation algorithm for objects with very low edge contrast. In Casasent DP, Hall EL, Roening J, editors, Proceedings of SPIE - The International Society for Optical Engineering. Vol. 5608. 2004. p. 162-168. 19 https://doi.org/10.1117/12.580136
McCarthy, Stephen ; Miller, Timothy C. ; Bankman, Isaac N. / Segmentation algorithm for objects with very low edge contrast. Proceedings of SPIE - The International Society for Optical Engineering. editor / D.P. Casasent ; E.L. Hall ; J. Roening. Vol. 5608 2004. pp. 162-168
@inproceedings{879e44cdc53641839f3f0eab42899bdd,
title = "Segmentation algorithm for objects with very low edge contrast",
abstract = "We present an algorithm for segmentation of objects with very low edge contrast, such as microcalcifications in mammogram images. Most methods used to segment microcalcifications have algorithmic aspects that could raise operational difficulties, such as thresholds or windows that must be selected manually, or parametric models of the data. The presented algorithm does not use any of these techniques and does not require that any parameters be set by a user. It builds upon an earlier algorithm presented in, but is much faster and also applicable to a wider range of objects to be segmented. The algorithm's approach is based on the extension of radial intensity profiles from a given seed point to the edge of the image. A first derivative analysis is used to find an edge point pixel along each directional intensity profile. These points are connected and the resulting object border is filled using a constrained dilatation operation to form a complete region. Results from the tested mammography images indicate that the segmented regions compare closely to those expected from visual inspection.",
keywords = "Automated image analysis system, Fast algorithm, Image segmentation, Low contrast image",
author = "Stephen McCarthy and Miller, {Timothy C.} and Bankman, {Isaac N.}",
year = "2004",
doi = "10.1117/12.580136",
language = "English (US)",
volume = "5608",
pages = "162--168",
editor = "D.P. Casasent and E.L. Hall and J. Roening",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

}

TY - GEN

T1 - Segmentation algorithm for objects with very low edge contrast

AU - McCarthy, Stephen

AU - Miller, Timothy C.

AU - Bankman, Isaac N.

PY - 2004

Y1 - 2004

N2 - We present an algorithm for segmentation of objects with very low edge contrast, such as microcalcifications in mammogram images. Most methods used to segment microcalcifications have algorithmic aspects that could raise operational difficulties, such as thresholds or windows that must be selected manually, or parametric models of the data. The presented algorithm does not use any of these techniques and does not require that any parameters be set by a user. It builds upon an earlier algorithm presented in, but is much faster and also applicable to a wider range of objects to be segmented. The algorithm's approach is based on the extension of radial intensity profiles from a given seed point to the edge of the image. A first derivative analysis is used to find an edge point pixel along each directional intensity profile. These points are connected and the resulting object border is filled using a constrained dilatation operation to form a complete region. Results from the tested mammography images indicate that the segmented regions compare closely to those expected from visual inspection.

AB - We present an algorithm for segmentation of objects with very low edge contrast, such as microcalcifications in mammogram images. Most methods used to segment microcalcifications have algorithmic aspects that could raise operational difficulties, such as thresholds or windows that must be selected manually, or parametric models of the data. The presented algorithm does not use any of these techniques and does not require that any parameters be set by a user. It builds upon an earlier algorithm presented in, but is much faster and also applicable to a wider range of objects to be segmented. The algorithm's approach is based on the extension of radial intensity profiles from a given seed point to the edge of the image. A first derivative analysis is used to find an edge point pixel along each directional intensity profile. These points are connected and the resulting object border is filled using a constrained dilatation operation to form a complete region. Results from the tested mammography images indicate that the segmented regions compare closely to those expected from visual inspection.

KW - Automated image analysis system

KW - Fast algorithm

KW - Image segmentation

KW - Low contrast image

UR - http://www.scopus.com/inward/record.url?scp=17644427344&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=17644427344&partnerID=8YFLogxK

U2 - 10.1117/12.580136

DO - 10.1117/12.580136

M3 - Conference contribution

AN - SCOPUS:17644427344

VL - 5608

SP - 162

EP - 168

BT - Proceedings of SPIE - The International Society for Optical Engineering

A2 - Casasent, D.P.

A2 - Hall, E.L.

A2 - Roening, J.

ER -