On the impact of lossy compression on hyperspectral image classification and unmixing

Fernando García-Vílchez, Jordi Muñoz-Marí, Maciel Zortea, Ian Blanes, Vicente González-Ruiz, Gustavo Camps-Valls, Antonio Plaza, Joan Serra-Sagristà

Research output: Contribution to journalArticleResearchpeer-review

84 Citations (Scopus)

Abstract

Hyperspectral data lossy compression has not yet achieved global acceptance in the remote sensing community, mainly because it is generally perceived that using compressed images may affect the results of posterior processing stages. This possible negative effect, however, has not been accurately characterized so far. In this letter, we quantify the impact of lossy compression on two standard approaches for hyperspectral data exploitation: spectral unmixing, and supervised classification using support vector machines. Our experimental assessment reveals that different stages of the linear spectral unmixing chain exhibit different sensitivities to lossy data compression. We have also observed that, for certain compression techniques, a higher compression ratio may lead to more accurate classification results. Even though these results may seem counterintuitive, this work explains these observations in light of the spatial regularization and/or whitening that most compression techniques perform and further provides recommendations on best practices when applying lossy compression prior to hyperspectral data classification and/or unmixing. © 2006 IEEE.
Original languageEnglish
Article number5570893
Pages (from-to)253-257
JournalIEEE Geoscience and Remote Sensing Letters
Volume8
Issue number2
DOIs
Publication statusPublished - 1 Mar 2011

Keywords

  • Endmember extraction
  • hyperspectral data lossy compression
  • image classification
  • linear spectral unmixing
  • principal component analysis (PCA)
  • regularization
  • support vector machine (SVM)
  • transform coding
  • wavelet transform

Fingerprint

Dive into the research topics of 'On the impact of lossy compression on hyperspectral image classification and unmixing'. Together they form a unique fingerprint.

Cite this