Face detection in color images based on skin color models

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

10 Citations (Scopus)

Abstract

Face finding is a very important initial step towards building up a fully automated face recognition system. Face detection by detecting skin like colors can achieve a high detection rate. In this paper, we presented a novel algorithm for face detection in color images with complex backgrounds. First a parallel structure for skin color detection is proposed to improve the accuracy of detections. The concept of the probability image has been introduced to utilize the color information in the traditional face detection methods specific for gray-scale images. After that, a classifier obtained from Adaboost training is applied to the result of skin detection to reduce the false positives. An experiment has been implemented to verify the improvement of this proposed research. And the proposed approach achieved a better result in this experiment.

Original languageEnglish
Title of host publicationIEEE Region 10 Annual International Conference, Proceedings/TENCON
Pages681-686
Number of pages6
DOIs
Publication statusPublished - 2010
Event2010 IEEE Region 10 Conference, TENCON 2010 - Fukuoka
Duration: 2010 Nov 212010 Nov 24

Other

Other2010 IEEE Region 10 Conference, TENCON 2010
CityFukuoka
Period10/11/2110/11/24

    Fingerprint

Keywords

  • Adaboost
  • Face detection
  • Face skin categorization
  • GMM
  • Skin color

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Science Applications

Cite this

Zou, L., & Kamata, S. (2010). Face detection in color images based on skin color models. In IEEE Region 10 Annual International Conference, Proceedings/TENCON (pp. 681-686). [5686631] https://doi.org/10.1109/TENCON.2010.5686631