Résumé

Face detection using Zernike moments is presented in this work. Face detection is obtained in two steps. In the first step the image to be treated is presented to a program which calculate its Zernike moments vector with appropriate parameters. In the second step, this reduced vector is used as an entry on the input layer of a neural network, beforehand trained with similar vectors. The neural network provides on its output layer a vector of co-ordinates in (R,?) representing pixels surrounding the face contained in the treated image.
The Zernike moments are used here for there properties of orthogonality and rotational invariability. The
experimental results of the application of our method on images of the XM2VTS database are presented using a
new objective criteria for measuring performances.

Information

Item Type:	Journal
Divisions:	» Faculté des Sciences et de la Technologie
ePrint ID:	35
Date Deposited:	2014-10-30
Further Information:	Google Scholar
URI:	https://www.univ-soukahras.dz/fr/publication/article/35

BibTex