Academic Journals Database
Disseminating quality controlled scientific knowledge

Personal Identification with F ace Bio metrics using Co lor Local Texture Features

ADD TO MY LIST
 
Author(s): Vani A.Hiremani | Sanjeevakumar | M.Hatture

Journal: International Journal of Advanced Computer Research
ISSN 2249-7277

Volume: 3;
Issue: 10;
Start page: 138;
Date: 2013;
VIEW PDF   PDF DOWNLOAD PDF   Download PDF Original page

Keywords: Color local texture features | Local binary pat tern | Color space conversion | V i o la and Jones algorithm | Match score level | Distance based classifier

ABSTRACT
Facerecognition (FR) has received a significantinterest in pattern recognition and computer visiondue to the wide range of applications includingvideo surveillance, biometric identification, and faceindexing in multimedia contents. Recently, localtexturefeatures have gained reputation as powerfulface descriptors because they are believed to bemore robust to variations of facial pose, expression,occlusion, etc. In particular,local binary pattern(LBP)texture feature hasproven to be highlydiscriminative for FR due to different levels oflocality. Hence, it is proposedto employ thesefeatures along withcolor local texture feature forefficient FR system.The personal identificationaccuracy with face modality using color localtexture featuresis around 97% is achieved
Why do you need a reservation system?      Affiliate Program