A class of Robust Estimates for Detection in Hyperspectral Images using Elliptical Distributions Backgroung

Joana Frontera-Pons 1 Mélanie Mahot 1 Jean-Philippe Ovarlez 1, 2 Frédéric Pascal 1 Sze Kim Pang 3 Jocelyn Chanussot 4
1 SONDRA - Sondra, CentraleSupélec, Université Paris-Saclay
2 ONERA - The French Aerospace Lab [Palaiseau]
3 DSO National Laboratories
4 GIPSA-SIGMAPHY - SIGMAPHY
GIPSA-DIS - Département Images et Signal
Abstract : When dealing with impulsive background echoes, Gaussian model is no longer pertinent. We study in this paper the class of elliptically contoured (EC) distributions. They provide a multivariate location-scatter family of distributions that primarily serve as long tailed alternatives to the multivariate normal model. They are proven to represent a more accurate characterization of HSI data than models based on the multivariate Gaussian assumption. For data in ℝk, robust proposals for the sample covariance estimate are the M-estimators. We have also analyzed the performance of an adaptive non- Gaussian detector built with these improved estimators. Constant False Alarm Rate (CFAR) is pursued to allow the detector independence of nuisance parameters and false alarm regulation.
Document type :
Conference papers
Complete list of metadatas
https://hal-supelec.archives-ouvertes.fr/hal-00781727
Contributor : Anne-Hélène Picot <>
Submitted on : Monday, January 28, 2013 - 11:35:14 AM
Last modification on : Friday, June 21, 2019 - 11:18:04 AM

Identifiers

Collections

SUP_SONDRA | SUPELEC | SUP_E3S | ONERA | GIPSA-SIGMAPHY | GIPSA-DIS | GIPSA | UGA | UGA-TEST-TER

Citation

Joana Frontera-Pons, Mélanie Mahot, Jean-Philippe Ovarlez, Frédéric Pascal, Sze Kim Pang, et al.. A class of Robust Estimates for Detection in Hyperspectral Images using Elliptical Distributions Backgroung. 2012 IEEE International Geoscience and Remote Sensing Symposium (IGARSS 2012), Jul 2012, Munich, Germany. pp.4166 - 4169, ⟨10.1109/IGARSS.2012.6350938⟩. ⟨hal-00781727⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

505

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD