Information theoretical based feature selection approach for human skin detection

K. S. Chenaoua, Ahmed Bouridane

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

2 Citations (Scopus)

Abstract

Detection of human skin in colored images has always been performed in known standard color spaces. In this paper a new color space coordinate is proposed based on popular existing color spaces but taking into account the most representative ones. Selection of the best color components is based on the use of the mutual information and maximum relevance minimum redundancy technique. A Gaussian model based classifier is used to test the performance of the proposed color space transformation.
Original languageEnglish
Title of host publicationProceedings of the 2010 IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP)
Place of PublicationPiscataway, NJ
PublisherIEEE
Pages1122-1125
ISBN (Print)978-1424442959
Publication statusPublished - 17 Mar 2010
EventICASSP International Conference on Audio, Speech and Signal Processing - Dallas, Texas, USA
Duration: 17 Mar 2010 → …

Conference

ConferenceICASSP International Conference on Audio, Speech and Signal Processing
Period17/03/10 → …

Fingerprint Dive into the research topics of 'Information theoretical based feature selection approach for human skin detection'. Together they form a unique fingerprint.

Cite this