Face recognition using relevance
weighted LDA with QR decomposition
Khalid Chougdali, Mohamed Jedra, Noureddine Zahid
Laboratory of Conception and System, Faculty of Science
Avenue Ibn Batouta, B.P.1014, Rabat, Morocco
Abstract:
We present a new variant on Linear Discriminant Analysis (LDA)
for face recognition. This algorithm combines the advantages
of the recent LDA enhancements namely relevance weighted LDA
and QR decomposition matrix analysis. Experiments on two
well known facial databases show the effectiveness of the
proposed method. Comparisons with other LDA-based methods
show that our method improves the LDA classification
performance.
Keywords:
Face recognition, Linear discriminant analysis,
relevance weighted LDA, LDA/QR, QR
decomposition.
Biographies:
 |
Khalid Chougdali,
was born in Marrakech, Morocoo in 1977. He graduated
from the Faculty of science, university Caddi Ayad,
Marrakech, and received the M.S degree in computer
science in 2001. Currently, he is preparing his
Doctorat National degree in computer science. His
research interests are biometrics, pattern
recognition, machine intelligence and image
processing. |
 |
Mohamed Jedra,
received the « Doctorat de Troisième Cycle » and « Doctorat
d’Etat » degrees in Electronics Engineering and
Informatics; respectively in 1990 and 1999, from
Mohammed V University in Rabat, Morocco. From 1990
to 1999, he was the Network and Internet Center
Manager in the Faculty of Sciences and Assistant
Professor at the Department of Physics. In 1999, he
became a Professor Habilité of Informatics in the
same Department and in 2003 he was promoted to the
position of Professor. From 1987 to the present, he
was a member of the Conception & Systems Laboratory.
He is the co-founder and the current Chair of the
Architecture of Informatic Systems UFR / Formation
and Research Unit in the Faculty of Sciences in
Rabat since 2003. His main research interests
include computational intelligence, pattern
recognition and biometric. He was a reviewer of
several conferences dealing with topics related to
computational intelligence and pattern recognition.
He has published numerous refereed papers in
specialised journals and conferences. Mohamed Jedra
is a member of IEEE since 1995. He is also a member
of IEEE Computer Society, IEEE Computational
Intelligence Society and IEEE Signal Processing
Society. |
 |
Noureddine Zahid,
holds « Doctorat de Troisième Cycle » and « Doctorat
d’Etat » degrees in Electronics Engineering and
Informatics; all form the Mohammed V University in
Rabat, Morocco. He was an Assistant Professor in
computer science at the Department of Physics
between 1988 and 1999, and Professor “Habilité”
between 1999 and 2003. In 2003, he was promoted to
the position of Professor in the same department. He
is a member of Conception & Systems Laboratory. He
is the co-founder and the current Chair of the
Informatics Telecommunications and Imagery master (ITI)
in the Faculty of Science in Rabat. He has presented
and published many articles in scientific journals
and conferences. He has reviewed numerous papers for
international journals in the field of fuzzy systems
and pattern recognition. His research interests lie
in the areas of fuzzy logic, pattern recognition,
machine learning and biometric. |
BibTex:
@ARTICLE{P1150627006,
AUTHOR = {Khalid Chougdali and Mohamed
Jedra and Noureddine Zahid},
TITLE = {Face recognition
using relevance weighted LDA with QR decomposition},
JOURNAL ={ICGST International Journal on Graphics, Vision and Image
Processing},
YEAR = {2006},
VOLUME = {6},
ISSUE ={2},
PAGES={27--31}
}
( Full Paper
458KB)
|
