Integration of contextual knowledge in unsupervised sub-pixel classification

P. V. Arun, Krishna Mohan Buddhiraju, Alok Porwal

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

4 Scopus citations

Abstract

In this paper, we investigate the use of coarse image features for predicting class label distributions at a finer scale. The major contributions of this work are 1) use of coarse image features to improve the optimization formulation of conventional rank based approaches 2) use of inter class compatibility information from coarse images to refine the predicted target distribution 3) an enhanced unsupervised variogram based sub-pixel mapping approach 4) inclusion of abundance estimation uncertainty in the unmixing process. The proposed modifications on rank based and variogram based approaches have produced an accuracy improvement of 10-15%. The sensitivities of these approaches towards tunable parameters are also analyzed.

Original languageEnglish
Title of host publication2016 8th Workshop on Hyperspectral Image and Signal Processing
Subtitle of host publicationEvolution in Remote Sensing, WHISPERS 2016
PublisherInstitute of Electrical and Electronics Engineers
ISBN (Electronic)9781509006083
DOIs
StatePublished - 28 Jun 2016
Externally publishedYes
Event8th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, WHISPERS 2016 - Los Angeles, United States
Duration: 21 Aug 201624 Aug 2016

Publication series

NameWorkshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing
Volume0
ISSN (Print)2158-6276

Conference

Conference8th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, WHISPERS 2016
Country/TerritoryUnited States
CityLos Angeles
Period21/08/1624/08/16

Keywords

  • Context
  • Sub-pixel mapping
  • Variogram

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing

Fingerprint

Dive into the research topics of 'Integration of contextual knowledge in unsupervised sub-pixel classification'. Together they form a unique fingerprint.

Cite this