Classification of hyperspectral data from urban areas based on extended morphological profiles

Jón Atli Benediktsson*, Jón Aevar Palmason, Johannes R. Sveinsson

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

982 Citations (Scopus)

Abstract

Classification of hyperspectral data with high spatial resolution from urban areas is investigated. A method based on mathematical morphology for preprocessing of the hyperspectral data is proposed. In this approach, opening and closing morphological transforms are used in order to isolate bright (opening) and dark (closing) structures in images, where bright/dark means brighter/darker than the surrounding features in the images. A morphological profile is constructed based on the repeated use of openings and closings with a structuring element of increasing size, starting with one original image. In order to apply the morphological approach to hyperspectral data, principal components of the hyperspectral imagery are computed. The most significant principal components are used as base images for an extended morphological profile, i.e., a profile based on more than one original image. In experiments, two hyperspectral urban datasets are classified. The proposed method is used as a preprocessing method for a neural network classifier and compared to more conventional classification methods with different types of statistical computations and feature extraction.

Original languageEnglish
Pages (from-to)480-491
Number of pages12
JournalIEEE Transactions on Geoscience and Remote Sensing
Volume43
Issue number3
DOIs
Publication statusPublished - Mar 2005

Bibliographical note

Funding Information:
Manuscript received April 7, 2004; revised September 3, 2004. This work was supported in part by the Icelandic Research Council and in part by the Research Fund of the University of Iceland. An early version of this paper was presented at the IEEE Workshop on Advances in Techniques for Analysis of Remotely Sensed Data—A Workshop Honoring Prof. David A. Landgrebe, Oct. 27–28, NASA Goddard Visitor Center, Washington, DC, USA, 2003.

Funding Information:
The data in this experiment are very fine resolution hyper-spectral data, part of the records of four flight lines over the urban area of Pavia, in northern Italy [13]. The flight was done in the framework of the HySens project, managed by Deutsches Zentrum fuer Luft-und Raumfahrt (the German Aerospace Center) and sponsored by the European Union within the transnational access to major research infrastructures (Contract no. HPRI-CT-1999-00075). The urban area was imaged by means of the Digital Airborne Imaging Spectrometer (DAIS). The flight altitude was chosen as the lowest available for the airplane, which resulted in a spatial resolution of 2.6 m. The lines were chosen so that the images are partially overlapping,

Other keywords

  • Hyperspectral remote sensing data
  • Morphological profiles
  • Neural networks
  • Principal components

Fingerprint

Dive into the research topics of 'Classification of hyperspectral data from urban areas based on extended morphological profiles'. Together they form a unique fingerprint.

Cite this