To be or not to be convex? A study on regularization in hyperspectral image classification

Devis Tuia, Remi Flamary, Michel Barlaud

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

4 Citations (Scopus)

Abstract

Hyperspectral image classification has long been dominated by convex models, which provide accurate decision functions exploiting all the features in the input space. However, the need for high geometrical details, which are often satisfied by using spatial filters, and the need for compact models (i.e. relying on models issued form reduced input spaces) has pushed research to study alternatives such as sparsity inducing regularization, which promotes models using only a subset of the input features. Although successful in reducing the number of active inputs, these models can be biased and sometimes offer sparsity at the cost of reduced accuracy. In this paper, we study the possibility of using non-convex regularization, which limits the bias induced by the regularization. We present and compare four regularizers, and then apply them to hyperspectral classification with different cost functions.

Original languageEnglish
Title of host publication2015 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages4947-4950
Number of pages4
ISBN (Print)9781479979295
DOIs
Publication statusPublished - 10 Nov 2015
Externally publishedYes
EventIEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Milan, Italy
Duration: 26 Jul 201531 Jul 2015

Publication series

NameInternational Geoscience and Remote Sensing Symposium (IGARSS)
Volume2015-November

Conference

ConferenceIEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015
Country/TerritoryItaly
CityMilan
Period26/07/1531/07/15

Fingerprint

Dive into the research topics of 'To be or not to be convex? A study on regularization in hyperspectral image classification'. Together they form a unique fingerprint.

Cite this