A fast connected components labeling algorithm and its application to real-time pupil detection

Prasad Gabbur, Hong Hua, Jacobus J Barnard

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

We describe a fast connected components labeling algorithm using a region coloring approach. It computes region attributes such as size, moments, and bounding boxes in a single pass through the image. Working in the context of real-time pupil detection for an eye tracking system, we compare the time performance of our algorithm with a contour tracing-based labeling approach and a region coloring method developed for a hardware eye detection system. We find that region attribute extraction performance exceeds that of these comparison methods. Further, labeling each pixel, which requires a second pass through the image, has comparable performance.

Original languageEnglish (US)
Pages (from-to)779-787
Number of pages9
JournalMachine Vision and Applications
Volume21
Issue number5
DOIs
StatePublished - Aug 2010

Fingerprint

Labeling
Coloring
Pixels
Hardware

Keywords

  • Connected components labeling
  • Eye tracking
  • Pupil detection
  • Region coloring
  • Segmentation

ASJC Scopus subject areas

  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Software
  • Computer Science Applications

Cite this

A fast connected components labeling algorithm and its application to real-time pupil detection. / Gabbur, Prasad; Hua, Hong; Barnard, Jacobus J.

In: Machine Vision and Applications, Vol. 21, No. 5, 08.2010, p. 779-787.

Research output: Contribution to journalArticle

@article{1307edac32c04a11af1e0add78c30351,
title = "A fast connected components labeling algorithm and its application to real-time pupil detection",
abstract = "We describe a fast connected components labeling algorithm using a region coloring approach. It computes region attributes such as size, moments, and bounding boxes in a single pass through the image. Working in the context of real-time pupil detection for an eye tracking system, we compare the time performance of our algorithm with a contour tracing-based labeling approach and a region coloring method developed for a hardware eye detection system. We find that region attribute extraction performance exceeds that of these comparison methods. Further, labeling each pixel, which requires a second pass through the image, has comparable performance.",
keywords = "Connected components labeling, Eye tracking, Pupil detection, Region coloring, Segmentation",
author = "Prasad Gabbur and Hong Hua and Barnard, {Jacobus J}",
year = "2010",
month = "8",
doi = "10.1007/s00138-009-0183-1",
language = "English (US)",
volume = "21",
pages = "779--787",
journal = "Machine Vision and Applications",
issn = "0932-8092",
publisher = "Springer Verlag",
number = "5",

}

TY - JOUR

T1 - A fast connected components labeling algorithm and its application to real-time pupil detection

AU - Gabbur, Prasad

AU - Hua, Hong

AU - Barnard, Jacobus J

PY - 2010/8

Y1 - 2010/8

N2 - We describe a fast connected components labeling algorithm using a region coloring approach. It computes region attributes such as size, moments, and bounding boxes in a single pass through the image. Working in the context of real-time pupil detection for an eye tracking system, we compare the time performance of our algorithm with a contour tracing-based labeling approach and a region coloring method developed for a hardware eye detection system. We find that region attribute extraction performance exceeds that of these comparison methods. Further, labeling each pixel, which requires a second pass through the image, has comparable performance.

AB - We describe a fast connected components labeling algorithm using a region coloring approach. It computes region attributes such as size, moments, and bounding boxes in a single pass through the image. Working in the context of real-time pupil detection for an eye tracking system, we compare the time performance of our algorithm with a contour tracing-based labeling approach and a region coloring method developed for a hardware eye detection system. We find that region attribute extraction performance exceeds that of these comparison methods. Further, labeling each pixel, which requires a second pass through the image, has comparable performance.

KW - Connected components labeling

KW - Eye tracking

KW - Pupil detection

KW - Region coloring

KW - Segmentation

UR - http://www.scopus.com/inward/record.url?scp=77956577596&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77956577596&partnerID=8YFLogxK

U2 - 10.1007/s00138-009-0183-1

DO - 10.1007/s00138-009-0183-1

M3 - Article

AN - SCOPUS:77956577596

VL - 21

SP - 779

EP - 787

JO - Machine Vision and Applications

JF - Machine Vision and Applications

SN - 0932-8092

IS - 5

ER -