Please use this identifier to cite or link to this item:
Title: Domain adaptation for polsar land classification using linear discriminative laplacian eigenmaps
Authors: Sun, W
Li, P
Yang, J
Shi, L 
Zhao, L
Keywords: Dimensionality reduction
Domain adaptation
Supervised classification
Issue Date: 2017
Publisher: Institute of Electrical and Electronics Engineers
Source: International Geoscience and Remote Sensing Symposium (IGARSS), 23-28 July 2017, 8127697, p. 3278-3281 How to cite?
Abstract: Recently, with the rapid development of earth observation (EO) techniques, the similar object information has been acquired in different regions, by the use of various sensors. It brings up a new challenge, that is, how to identify cross-domain objects. To cope with this difficulty, we first revisit the linear discriminative Laplacian eigenmaps (LDLE) in this paper, and further add a Bregman divergence (BD) based regularization term into it. The experiment results demonstrate that, the combination of LDLE and BD can learn a good linear transformation of polarimetric synthetic aperture radar (PolSAR) data and improve classification performances.
ISBN: 9.78151E+12
DOI: 10.1109/IGARSS.2017.8127697
Appears in Collections:Conference Paper

View full-text via PolyU eLinks SFX Query
Show full item record

Google ScholarTM



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.