Vol. 16, No. 1, 21-28, 2017

Atmospheric correction of imaging spectroscopy data using shadow-based quantification of aerosol scattering effects
Daniel Schläpfer, and Rudolf Richter

The atmospheric correction accuracy strongly depends on the correct estimate of the aerosol scattering effects. This paper shows first results of a new aerosol optical thickness inversion method and its use for improved atmospheric correction of high spatial resolution imaging spectroscopy data. The approach uses small scale shadow pixels for the determination of the atmospheric scattering by inverting the shadow correction within the ATCOR® atmospheric compensation method. The detection of shadow pixels is done by a blue to red ratio which is further adjusted by the near infrared band in order to take the vegetation bias into account. On high resolution instruments with resolutions below 5 m, a decent quantity of shaded pixels can be found by this method in a reliable way. Using this shadow mask, the aerosol inversion is done. The aerosol contents are varied in a way that retrieves a correction of shaded areas to a brightness comparable to non-shaded areas, leading to the best fitting aerosol amount. The shadow based aerosol optical thickness (SHAOT) method is tested on APEX and HYSPEX airborne imaging spectroscopy data. It can be shown that the atmospheric compensation using the such derived aerosol contents are significantly improved in comparison to standard correction techniques.

View Full Text (pdf file, 5 MB) previous page
DOI: 10.12760/01-2017-1-03

Submitted: 07 May 2017
Revised: 07 Oct 2017
Accepted: 30 Oct 2017
Published: 02 Dec 2017
Responsible editor: Rainer Reuter

Schläpfer D & R Richter, 2017. Atmospheric correction of imaging spectroscopy data using shadow-based quantification of aerosol scattering effects. EARSeL eProceedings, 16(1): 21-28

EARSeL European Association of Remote Sensing Laboratories, Strasbourg, France


BIS Library and Information System, Carl von Ossietzky University of Oldenburg


Indexed in Scopus

DOAJ logo

Directory of Open Access Journals


Opening access to research

JournalGuide logo

Find the best journal for your research

Creative Commons License

ISSN 1729-3782