2002ePrints Soton (University of Southampton)Open access

The role of field spectroscopy in airborne sensor calibration: the example of the NERC CASI

E. Rollin, E.J. Milton, Karen Anderson

Open full text 2 citations

Abstract

This paper reports on a preliminary experiment to undertake in-flight radiance calibration validation (cal-val) of the NERC CASI. The approach involved comparing CASI radiance values of a single, well characterised target with synchronised measurements of target radiance obtained at the ground with a calibrated spectroradiometer. CASI data were obtained at two altitudes: 853m and 3109m, in spatial mode using the default vegetation band-set. For the lower altitude image, there was very good agreement between the image and ground-measured radiance especially in bands 4 to 6 (670nm-710nm). An atmospheric correction based on the multi-height image data significantly improved the image versus ground comparison for bands 1 to 3, where scattering dominates. However, the correction performed less well for the two NIR bands coincident with the atmospheric absorption regions (band 9 at 762nm and band 13 at 940nm). For the asphalt target, estimating radiance from spectral reflectance measured post-flight combined with spectral irradiance measured during the flights was found to be a viable alternative to measuring radiance. Overall, the results suggest the method could be developed further to provide in-flight calibration validation of the CASI to supplement or replace the mid-season laboratory calibration.

Open-access reader

About this research paper

What this paper is about

This paper reports on a preliminary experiment to undertake in-flight radiance calibration validation (cal-val) of the NERC CASI. The approach involved comparing CASI radiance values of a single, well characterised target with synchronised measurements of target radiance obtained at the ground with a calibrated spectroradiometer. CASI data were obtained at two altitudes: 853m and 3109m, in spatial mode using the default vegetation band-set. For the lower altitude image, there was very good agreement between the image and ground-measured radiance especially in bands 4 to 6 (670nm-710nm). An atmospheric correction based on the multi-height image data significantly improved the image versus ground comparison for bands 1 to 3, where scattering dominates. However, the correction performed less well for the two NIR bands coincident with the atmospheric absorption regions (band 9 at 762nm and band 13 at 940nm). For the asphalt target, estimating radiance from spectral reflectance measured post-flight combined with spectral irradiance measured during the flights was found to be a viable alternative to measuring radiance. Overall, the results suggest the method could be developed further to provide in-flight calibration validation of the CASI to supplement or replace the mid-season laboratory calibration.

Why it matters

OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

This paper reports on a preliminary experiment to undertake in-flight radiance calibration validation (cal-val) of the NERC CASI. The approach involved comparing CASI radiance values of a single, well characterised target with synchronised measurements of target radiance obtained at the ground with a calibrated spectroradiometer. CASI data were obtained at two altitudes: 853m and 3109m, in spatial mode using the default vegetation band-set. For the lower altitude image, there was very good agreement between the image and ground-measured radiance especially in bands 4 to 6 (670nm-710nm). An atmospheric correction based on the multi-height image data significantly improved the image versus ground comparison for bands 1 to 3, where scattering dominates. However, the correction performed less well for the two NIR bands coincident with the atmospheric absorption regions (band 9 at 762nm and band 13 at 940nm). For the asphalt target, estimating radiance from spectral reflectance measured post-flight combined with spectral irradiance measured during the flights was found to be a viable alternative to measuring radiance. Overall, the results suggest the method could be developed further to provide in-flight calibration validation of the CASI to supplement or replace the mid-season laboratory calibration.

Key concepts: Radiance, Spectroradiometer, Remote sensing, Irradiance, Calibration, Environmental science, Atmospheric correction, Spectral bands

Related papers

Back to paper searchBrowse research topicsOriginal source
The role of field spectroscopy in airborne sensor calibration: the example of the NERC CASI — Research Paper | ScholarLens