Szczegóły publikacji
Opis bibliograficzny
Image segmentation with a hybrid ensemble of one-class support vector machines / Bogusław CYGANEK // W: Hybrid Artificial Intelligence Systems : 5th international conference, HAIS 2010 : San Sebastián, Spain, June 23–25, 2010 : proceedings , Pt. 1 / eds. Manuel Graña Romay, Emilio Corchado, M. Teresa Garcia-Sebastian. — Berlin ; Heidelberg : Springer-Verlag, cop. 2010. — ( Lecture Notes in Computer Science ; ISSN 0302-9743 ; LNCS 6076. Lecture Notes in Artificial Intelligence ). — ISBN: 978-3-642-13768-6; ISBN: 3-642-13768-7; e-ISBN: 978-3-642-13769-3. — S. 254–261. — Bibliogr. s. 261, Abstr.
Autor
Dane bibliometryczne
| ID BaDAP | 53002 |
|---|---|
| Data dodania do BaDAP | 2010-08-30 |
| DOI | 10.1007/978-3-642-13769-3_31 |
| Rok publikacji | 2010 |
| Typ publikacji | materiały konferencyjne (aut.) |
| Otwarty dostęp |  |
| Czasopismo/seria | Lecture Notes in Computer Science |
Abstract
In this paper an efficient method of image segmentation from large data samples is presented. Segmentation is stated as a novelty detection problem for which the one-class support vector machines (OC-SVM) are employed. However, to improve performance and scalability the input space of samples is first k-means partitioned, and then each partition is independently trained with an OC-SVM. This way a parallel structure of expert classifiers is obtained with of a small average number of support vectors and high precision. © 2010 Springer-Verlag.
