Target detection in inhomogeneous non-gaussian hyperspectral data, based on non-parametric density estimation

G. A. Tidhar, S. R. Rotman

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

6 Scopus citations

Abstract

Performance of algorithms for target signal detection in Hyperspectral Imagery (HSI) is often deteriorated when the data is neither statistically homogeneous nor Gaussian or when its Joint Probability Density (JPD) does not match any presumed particular parametric model. In this paper we propose a novel detection algorithm which first attempts at dividing data domain into mostly Gaussian and mostly Non-Gaussian (NG) subspaces, and then estimates the JPD of the NG subspace with a non-parametric Graph-based estimator. It then combines commonly used detection algorithms operating on the mostly-Gaussian sub-space and an LRT calculated directly with the estimated JPD of the NG sub-space, to detect anomalies and known additive-type target signals. The algorithm performance is compared to commonly used algorithms and is found to be superior in some important cases.

Original languageEnglish
Title of host publicationAlgorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XIX
DOIs
StatePublished - 12 Aug 2013
EventAlgorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XIX - Baltimore, MD, United States
Duration: 29 Apr 20132 May 2013

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume8743
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

ConferenceAlgorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XIX
Country/TerritoryUnited States
CityBaltimore, MD
Period29/04/132/05/13

Keywords

  • Density-estimation
  • Graph
  • Hyperspectral
  • Non-parametric
  • Target-detection

Fingerprint

Dive into the research topics of 'Target detection in inhomogeneous non-gaussian hyperspectral data, based on non-parametric density estimation'. Together they form a unique fingerprint.

Cite this