Pre-attentive detection of perceptually important regions in facial images

Alexander Golovan, Myung Hyun Yoo, Seong Whan Lee

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Cascade method, a new way of calculating orientation components in an image, was developed to extract some important regions from a given image of human face. By combining local orientation components and thresholding them, we construct five feature maps and a final composite map which is a linear combination of the five maps. As in human visual perception, the composite map operates like pre-attentive processing in early stage of vision, and then shows robustness in selecting the most informative areas of images. For 50 non-normalized face images from ORL 1 database, it showed 91% of detecting accuracy which is the ratio of corresponding points between the feature maps of whole image and same maps of important regions in that image such as eyes, nose and mouth, etc.

Original languageEnglish
Title of host publicationProceedings - International Conference on Pattern Recognition
Pages1092-1095
Number of pages4
Volume15
Edition1
Publication statusPublished - 2000

Fingerprint

Composite materials
Processing

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Hardware and Architecture
  • Electrical and Electronic Engineering

Cite this

Golovan, A., Yoo, M. H., & Lee, S. W. (2000). Pre-attentive detection of perceptually important regions in facial images. In Proceedings - International Conference on Pattern Recognition (1 ed., Vol. 15, pp. 1092-1095)

Pre-attentive detection of perceptually important regions in facial images. / Golovan, Alexander; Yoo, Myung Hyun; Lee, Seong Whan.

Proceedings - International Conference on Pattern Recognition. Vol. 15 1. ed. 2000. p. 1092-1095.

Research output: Chapter in Book/Report/Conference proceedingChapter

Golovan, A, Yoo, MH & Lee, SW 2000, Pre-attentive detection of perceptually important regions in facial images. in Proceedings - International Conference on Pattern Recognition. 1 edn, vol. 15, pp. 1092-1095.
Golovan A, Yoo MH, Lee SW. Pre-attentive detection of perceptually important regions in facial images. In Proceedings - International Conference on Pattern Recognition. 1 ed. Vol. 15. 2000. p. 1092-1095
Golovan, Alexander ; Yoo, Myung Hyun ; Lee, Seong Whan. / Pre-attentive detection of perceptually important regions in facial images. Proceedings - International Conference on Pattern Recognition. Vol. 15 1. ed. 2000. pp. 1092-1095
@inbook{5e20c6ef012749a4aef41ca76b8d98b0,
title = "Pre-attentive detection of perceptually important regions in facial images",
abstract = "Cascade method, a new way of calculating orientation components in an image, was developed to extract some important regions from a given image of human face. By combining local orientation components and thresholding them, we construct five feature maps and a final composite map which is a linear combination of the five maps. As in human visual perception, the composite map operates like pre-attentive processing in early stage of vision, and then shows robustness in selecting the most informative areas of images. For 50 non-normalized face images from ORL 1 database, it showed 91{\%} of detecting accuracy which is the ratio of corresponding points between the feature maps of whole image and same maps of important regions in that image such as eyes, nose and mouth, etc.",
author = "Alexander Golovan and Yoo, {Myung Hyun} and Lee, {Seong Whan}",
year = "2000",
language = "English",
volume = "15",
pages = "1092--1095",
booktitle = "Proceedings - International Conference on Pattern Recognition",
edition = "1",

}

TY - CHAP

T1 - Pre-attentive detection of perceptually important regions in facial images

AU - Golovan, Alexander

AU - Yoo, Myung Hyun

AU - Lee, Seong Whan

PY - 2000

Y1 - 2000

N2 - Cascade method, a new way of calculating orientation components in an image, was developed to extract some important regions from a given image of human face. By combining local orientation components and thresholding them, we construct five feature maps and a final composite map which is a linear combination of the five maps. As in human visual perception, the composite map operates like pre-attentive processing in early stage of vision, and then shows robustness in selecting the most informative areas of images. For 50 non-normalized face images from ORL 1 database, it showed 91% of detecting accuracy which is the ratio of corresponding points between the feature maps of whole image and same maps of important regions in that image such as eyes, nose and mouth, etc.

AB - Cascade method, a new way of calculating orientation components in an image, was developed to extract some important regions from a given image of human face. By combining local orientation components and thresholding them, we construct five feature maps and a final composite map which is a linear combination of the five maps. As in human visual perception, the composite map operates like pre-attentive processing in early stage of vision, and then shows robustness in selecting the most informative areas of images. For 50 non-normalized face images from ORL 1 database, it showed 91% of detecting accuracy which is the ratio of corresponding points between the feature maps of whole image and same maps of important regions in that image such as eyes, nose and mouth, etc.

UR - http://www.scopus.com/inward/record.url?scp=33750958304&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33750958304&partnerID=8YFLogxK

M3 - Chapter

VL - 15

SP - 1092

EP - 1095

BT - Proceedings - International Conference on Pattern Recognition

ER -