Skin colour-based face detection in colour images

Wen Hsiang Lai, Chang Tsun Li

Research output: Book chapter/Published conference paperConference paper

14 Citations (Scopus)

Abstract

We propose in this work a method for detecting faces in colour images with complex backgrounds. The approach starts with the transformation of the image pixels from the RGB colour space to the chrominance space (YCbCr). Secondly, a Gaussian model is fitted on the transformed image in order to calculate the likelihood of skin for each pixel and to create a likelihood image. Thirdly, by thresholding the likelihood image, skin pixels are segmented to form a binary skin map, which contains the candidate face regions. Finally, a verification process is carried out to determine whether these candidate face regions are real faces or not.

Original languageEnglish
Title of host publicationProceedings - IEEE International Conference on Video and Signal Based Surveillance 2006, AVSS 2006
DOIs
Publication statusPublished - 2006
EventIEEE International Conference on Video and Signal Based Surveillance 2006, AVSS 2006 - Sydney, NSW, Australia
Duration: 22 Nov 200624 Nov 2006

Conference

ConferenceIEEE International Conference on Video and Signal Based Surveillance 2006, AVSS 2006
CountryAustralia
CitySydney, NSW
Period22/11/0624/11/06

Fingerprint

Face recognition
Skin
Pixels
Color

Cite this

Lai, W. H., & Li, C. T. (2006). Skin colour-based face detection in colour images. In Proceedings - IEEE International Conference on Video and Signal Based Surveillance 2006, AVSS 2006 [4020715] https://doi.org/10.1109/AVSS.2006.105
Lai, Wen Hsiang ; Li, Chang Tsun. / Skin colour-based face detection in colour images. Proceedings - IEEE International Conference on Video and Signal Based Surveillance 2006, AVSS 2006. 2006.
@inproceedings{9e03435b81034740901abc45865e26dc,
title = "Skin colour-based face detection in colour images",
abstract = "We propose in this work a method for detecting faces in colour images with complex backgrounds. The approach starts with the transformation of the image pixels from the RGB colour space to the chrominance space (YCbCr). Secondly, a Gaussian model is fitted on the transformed image in order to calculate the likelihood of skin for each pixel and to create a likelihood image. Thirdly, by thresholding the likelihood image, skin pixels are segmented to form a binary skin map, which contains the candidate face regions. Finally, a verification process is carried out to determine whether these candidate face regions are real faces or not.",
keywords = "Biometrics, Face detection, Face recognition",
author = "Lai, {Wen Hsiang} and Li, {Chang Tsun}",
year = "2006",
doi = "10.1109/AVSS.2006.105",
language = "English",
isbn = "0769526888",
booktitle = "Proceedings - IEEE International Conference on Video and Signal Based Surveillance 2006, AVSS 2006",

}

Lai, WH & Li, CT 2006, Skin colour-based face detection in colour images. in Proceedings - IEEE International Conference on Video and Signal Based Surveillance 2006, AVSS 2006., 4020715, IEEE International Conference on Video and Signal Based Surveillance 2006, AVSS 2006, Sydney, NSW, Australia, 22/11/06. https://doi.org/10.1109/AVSS.2006.105

Skin colour-based face detection in colour images. / Lai, Wen Hsiang; Li, Chang Tsun.

Proceedings - IEEE International Conference on Video and Signal Based Surveillance 2006, AVSS 2006. 2006. 4020715.

Research output: Book chapter/Published conference paperConference paper

TY - GEN

T1 - Skin colour-based face detection in colour images

AU - Lai, Wen Hsiang

AU - Li, Chang Tsun

PY - 2006

Y1 - 2006

N2 - We propose in this work a method for detecting faces in colour images with complex backgrounds. The approach starts with the transformation of the image pixels from the RGB colour space to the chrominance space (YCbCr). Secondly, a Gaussian model is fitted on the transformed image in order to calculate the likelihood of skin for each pixel and to create a likelihood image. Thirdly, by thresholding the likelihood image, skin pixels are segmented to form a binary skin map, which contains the candidate face regions. Finally, a verification process is carried out to determine whether these candidate face regions are real faces or not.

AB - We propose in this work a method for detecting faces in colour images with complex backgrounds. The approach starts with the transformation of the image pixels from the RGB colour space to the chrominance space (YCbCr). Secondly, a Gaussian model is fitted on the transformed image in order to calculate the likelihood of skin for each pixel and to create a likelihood image. Thirdly, by thresholding the likelihood image, skin pixels are segmented to form a binary skin map, which contains the candidate face regions. Finally, a verification process is carried out to determine whether these candidate face regions are real faces or not.

KW - Biometrics

KW - Face detection

KW - Face recognition

UR - http://www.scopus.com/inward/record.url?scp=34547345924&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34547345924&partnerID=8YFLogxK

U2 - 10.1109/AVSS.2006.105

DO - 10.1109/AVSS.2006.105

M3 - Conference paper

AN - SCOPUS:34547345924

SN - 0769526888

SN - 9780769526881

BT - Proceedings - IEEE International Conference on Video and Signal Based Surveillance 2006, AVSS 2006

ER -

Lai WH, Li CT. Skin colour-based face detection in colour images. In Proceedings - IEEE International Conference on Video and Signal Based Surveillance 2006, AVSS 2006. 2006. 4020715 https://doi.org/10.1109/AVSS.2006.105