Publication detail
New Methods for Face Recognition
ŠŤASTNÝ, J. ŠKORPIL, V.
Original Title
New Methods for Face Recognition
Czech Title
Nové metody rezpoznání obličeje
English Title
New Methods for Face Recognition
Type
conference paper
Language
en
Original Abstract
The paper describes the PCA (Principal Component Analysis) method and its comparison with a method using a neural network with RBF (Radial Basis Function) for classification over PCA data. The PCA method has a very solid resolution power for a small database and in this paper a database sample is described that contains 4000 images for 40 different people. Using the PCA method is suitable.
Czech abstract
Příspěvek popisuje metodu PCA (Principal Component Analysis) a její porovnání s metodou používající RBF (Radial Basis Function) neuronovou síť pro klasifikaci PCA dat. Metoda PCA je výhodné řešení pro malé databáze a tento příspěvek popisuje databázi obsahující 4000 obrazů 40 rozdílných osob. Použití PCA je vhodné.
English abstract
The paper describes the PCA (Principal Component Analysis) method and its comparison with a method using a neural network with RBF (Radial Basis Function) for classification over PCA data. The PCA method has a very solid resolution power for a small database and in this paper a database sample is described that contains 4000 images for 40 different people. Using the PCA method is suitable.
Keywords
Algorithms, Image analysis, Neural networks, Object recognition
RIV year
2010
Released
20.08.2010
Publisher
Asszistencia
Location
Baden near Vienna
ISBN
978-963-88981-0-4
Book
Proceedings of 33th International Conference on Telecommunications and Signal Processing
Edition number
1
Pages from
156
Pages to
159
Pages count
4
BibTex
@inproceedings{BUT35609,
author="Jiří {Šťastný} and Vladislav {Škorpil}",
title="New Methods for Face Recognition",
annote="The paper describes the PCA (Principal Component Analysis) method and its comparison with a method using a neural network with RBF (Radial Basis Function) for classification over PCA data. The PCA method has a very solid resolution power for a small database and in this paper a database sample is described that contains 4000 images for 40 different people. Using the PCA method is suitable.",
address="Asszistencia",
booktitle="Proceedings of 33th International Conference on Telecommunications and Signal Processing",
chapter="35609",
howpublished="print",
institution="Asszistencia",
year="2010",
month="august",
pages="156--159",
publisher="Asszistencia",
type="conference paper"
}