Study of uncertainties of hyperspectral image based on Fourier waveform analysis

Author

Wu, Lun ; Gao, Zhenji ; Liu, Yu ; Yu, Hailong

Author_Institution

Inst. of Remote Sensing & Geographic Inf. Syst., Peking Univ.

Volume

5

fYear

2004

fDate

20-24 Sept. 2004

Firstpage

3279

Abstract

Satellite-based spectra are a complicated function of imaging time, imaging azimuth, imaging channel, atmospheric status and ground object category. This complicated function accordingly leads to the ambiguity and uncertainty of satellite-based spectra and image classification. The uncertainty of satellite-based spectra can be divided into two categories. One is background noise; the other is SODS (same object different spectrum) and DOSS (different object similar spectrum). A method to analyze and lower the DOSS uncertainty of satellite-based spectra based on Fourier waveform analysis is discussed in this paper

Keywords

Fourier transforms; geophysical signal processing; geophysical techniques; image classification; remote sensing; DOSS uncertainty; Fourier waveform analysis; SODS; atmospheric status; background noise; different object similar spectrum; gray value; ground object category; hyperspectral image uncertainties; image ambiguity; image classification; imaging azimuth; imaging channel; imaging time; same object different spectrum; satellite-based spectra; Atmospheric measurements; Atmospheric waves; Azimuth; Background noise; Geographic Information Systems; Hyperspectral imaging; Hyperspectral sensors; Image classification; Remote sensing; Uncertainty;

fLanguage

English

Publisher

ieee

Conference_Titel

Geoscience and Remote Sensing Symposium, 2004. IGARSS '04. Proceedings. 2004 IEEE International

Conference_Location

Anchorage, AK

Print_ISBN

0-7803-8742-2

Type

conf

DOI

10.1109/IGARSS.2004.1370402

Filename

1370402