文章基本信息

标题：Evaluation of One-class SVM for pixel-based and segment-based classification in remote sensing
本地全文：下载
作者：Andreas Braun
期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2010
卷号：XXXVIII - Part 3B
页码：160-165
出版社：Copernicus Publications
摘要：Despite the wide usage of standard cost support vector machines in remote sensing, a less common approach of SVM - named one-class SVM - has found only little attention. This seems surprising to us, as one-class SVM overcomes one important shortcoming of standard cost SVM: the problems occuring from unassigned classes present in the scene and the difficulties handeling the rejection class. We present an introduction into one-class SVM for classification in remote sensing and a comparison with standard cost SVM