A probabilistic neural network for face detection on segmented skin areas based on fuzzy rules

Anagnostopoulos, C; Anagnostopoulos, I; Vergados, D; Papaleonidopoulos, I; Kayaias, E; Loumos, V; Stasinopoulos, G

dc.contributor.author	Anagnostopoulos, C	en
dc.contributor.author	Anagnostopoulos, I	en
dc.contributor.author	Vergados, D	en
dc.contributor.author	Papaleonidopoulos, I	en
dc.contributor.author	Kayaias, E	en
dc.contributor.author	Loumos, V	en
dc.contributor.author	Stasinopoulos, G	en
dc.date.accessioned	2014-03-01T02:42:03Z
dc.date.available	2014-03-01T02:42:03Z
dc.date.issued	2002	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/30753
dc.subject	Artificial neural networks	en
dc.subject	Face detection	en
dc.subject	Fuzzy logic	en
dc.subject	Image processing	en
dc.subject.other	Cameras	en
dc.subject.other	Color image processing	en
dc.subject.other	Fuzzy sets	en
dc.subject.other	Image quality	en
dc.subject.other	Image segmentation	en
dc.subject.other	Neural networks	en
dc.subject.other	Probability	en
dc.subject.other	Digital camera	en
dc.subject.other	Face detection	en
dc.subject.other	Probabilistic neural network	en
dc.subject.other	Face recognition	en
dc.title	A probabilistic neural network for face detection on segmented skin areas based on fuzzy rules	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1109/MELECON.2002.1014642	en
heal.identifier.secondary	http://dx.doi.org/10.1109/MELECON.2002.1014642	en
heal.publicationDate	2002	en
heal.abstract	In this paper, the problem of face detection in color images is addressed. Unlike in face recognition, where the classes to be discriminated are different faces, in face detection, the two classes to be discriminated are ""face area"" and ""non-face area"". A novel approach to face detection is presented, which is based on a Probabilistic Neural Network (PNN) specially trained for the identification of faces from skin areas within the image plane, which were previously located using Fuzzy Logic rules. The performance of the whole system was tested using 137 color image scenes with various illumination conditions containing human faces. The images were taken with a common digital camera, at a resolution of 1280×960 pixels. The PNN coupled with the fuzzy logic rules became the basis for developing a computer-based face detection system, whose overall identification performance was 85.32%. One limitation of the system is that it is restricted only in detection of frontal human faces.	en
heal.journalName	Proceedings of the Mediterranean Electrotechnical Conference - MELECON	en
dc.identifier.doi	10.1109/MELECON.2002.1014642	en
dc.identifier.volume	1	en
dc.identifier.spage	493	en
dc.identifier.epage	497	en