title   
  

Yerel görünüm tabanlı yüz tanıma için değişik boyut indirme ve normalizasyon yöntemlerinin incelenmesi (Investigation of different dimension reduction and normalization methods for local appearance-based face recognition)

Topçu, Berkay and Erdoğan, Hakan (2009) Yerel görünüm tabanlı yüz tanıma için değişik boyut indirme ve normalizasyon yöntemlerinin incelenmesi (Investigation of different dimension reduction and normalization methods for local appearance-based face recognition). In: IEEE 17th Signal Processing and Communications Applications Conference, 2009 (SIU 2009), Antalya, Türkiye

[img]
Preview
PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
594Kb

Official URL: http://dx.doi.org/10.1109/SIU.2009.5136425

Abstract

Local appearance-based methods have been proposed recently for face recognition. We analyze the effects of different dimension reduction and normalization methods on local appearance-based face recognition in this paper. Each image is divided into equal sized blocks and six different dimension reduction methods are implemented for each block separately to create local visual feature vectors. On these local features, several normalization methods are applied in an attempt to eliminate the changes in lighting conditions and contrast differences among blocks of different face images. The experimental results show the improvements in recognition rates due to the effects of dimension reduction and normalization for three different classifiers. Usage of trainable dimension reduction methods instead of DCT and a new normalization method in our work (within-block normalization as referred in this paper) are two factors that makes difference from previous works in literature. The best performance is achieved using a block size of 16times16, performing dimension reduction using approximate pairwise accuracy criterion (aPAC) and applying within-block mean and variance normalization.

Item Type:Papers in Conference Proceedings
Subjects:T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5101-6720 Telecommunication
ID Code:12744
Deposited By:Hakan Erdoğan
Deposited On:18 Nov 2009 17:10
Last Modified:25 May 2011 14:19

Repository Staff Only: item control page