Skip to main navigation Skip to search Skip to main content

Joint Tensor Subspace Alignment on Multi-Angular Remote Sensing Image

  • School of Electronics and Information Engineering, Harbin Institute of Technology

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

High-resolution remote sensing satellites produce many images with different angles during the shooting process. Some of the angular images are labeled, but more are not labeled. How to use these labeled images to classify images of other different angles is a problem that needs to be solved. In view of this, we proposes an unsupervised DA approach by aligning tensor subspace and combining multi-angular images. The source subspace is aligned to the target subspace so that the source domain sample probability distribution is close to the target domain sample. Using images from different angles as source domains can make full use of their similarities and differences to achieve better classification results. To verify the effectiveness of this method, this paper conducts a set of five angular images taken by WorldView-2 satellites. The experimental results show that the proposed method can effectively distribute the source image group close to the target image distribution and achieve better classification.

Original languageEnglish
Title of host publication2018 9th Workshop on Hyperspectral Image and Signal Processing
Subtitle of host publicationEvolution in Remote Sensing, WHISPERS 2018
PublisherIEEE Computer Society
ISBN (Electronic)9781728115818
DOIs
StatePublished - Sep 2018
Externally publishedYes
Event9th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, WHISPERS 2018 - Amsterdam, Netherlands
Duration: 23 Sep 201826 Sep 2018

Publication series

NameWorkshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing
Volume2018-September
ISSN (Print)2158-6276

Conference

Conference9th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, WHISPERS 2018
Country/TerritoryNetherlands
CityAmsterdam
Period23/09/1826/09/18

Keywords

  • high resolution
  • joint classification
  • multi-angular
  • subspace alignment
  • tensor alignment

Fingerprint

Dive into the research topics of 'Joint Tensor Subspace Alignment on Multi-Angular Remote Sensing Image'. Together they form a unique fingerprint.

Cite this