Multipixel anomaly detection in noisy multispectral images

Research output: Contribution to journalArticlepeer-review

7 Scopus citations

Abstract

Basing ourselves on a novel segmentation algorithm for multispectral images, we consider how to detect multipixel anomalous objects in image cubes where spectral information is available. In particular, we have developed several morphological filters to compensate for noise that may be present in the initial cube. We show that for different types of noise (Gaussian and speckle), a modification of our morphology technique allows us to better detect targets without an enhanced false-alarm result.

Original languageEnglish
Article number023604
JournalOptical Engineering
Volume45
Issue number2
DOIs
StatePublished - 1 Feb 2006

Keywords

  • AVIRIS
  • Anomaly detection
  • Gaussian noise
  • Hyperspectral imaging
  • Morphological operation
  • Speckle noise
  • Spectral data

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics
  • General Engineering

Fingerprint

Dive into the research topics of 'Multipixel anomaly detection in noisy multispectral images'. Together they form a unique fingerprint.

Cite this