Automatic clustering of hyperspectral data

R. Salomon, S. Dolberg, S. R. Rotman

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

Abstract

The main goal of this research is to examine some new methods for the automatic clustering of hyperspectral data. Hyperspectral data consists of images which originate from the same physical phenomena at various wavelengths. Use of this data is common not only for medical processing but also for military purposes. In this unique research we will analyze hyperspectral data which has been taken of different types of events that evolve both temporally and spectrally. These events would seemingly be indistinguishable if only the spectral or the temporal dimensions were used. By exploiting the unique attributes of the hyperspectral temporal data, we show that we can significantly improve our target assignment capabilities. We will develop methods to evaluate our ability to correctly assign these events from each other. We will discuss how to automatically cluster such events and determine how many different types of events actually exist. Practical problems previously discussed in the literature will be demonstrated.

Original languageEnglish
Title of host publication2006 IEEE 24th Convention of Electrical and Electronics Engineers in Israel, IEEEI
PublisherInstitute of Electrical and Electronics Engineers
Pages330-333
Number of pages4
ISBN (Print)1424402301, 9781424402304
DOIs
StatePublished - 1 Jan 2006
Event2006 IEEE 24th Convention of Electrical and Electronics Engineers in Israel, IEEEI - Eilat, Israel
Duration: 15 Nov 200617 Nov 2006

Publication series

NameIEEE Convention of Electrical and Electronics Engineers in Israel, Proceedings

Conference

Conference2006 IEEE 24th Convention of Electrical and Electronics Engineers in Israel, IEEEI
Country/TerritoryIsrael
CityEilat
Period15/11/0617/11/06

Keywords

  • Clustering
  • Detonation
  • Hyperspectral data

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Electronic, Optical and Magnetic Materials

Fingerprint

Dive into the research topics of 'Automatic clustering of hyperspectral data'. Together they form a unique fingerprint.

Cite this