The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Publications Copernicus
Articles | Volume XL-7
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XL-7, 51–53, 2014
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XL-7, 51–53, 2014

  19 Sep 2014

19 Sep 2014

Practical example for use of the supervised vicarious calibration (SVC) method on multisource hyperspectral imagery data – ValCalHyp airborne hyperspectral campaign under the EUFAR framework

A. Brook1 and E. Ben Dor2 A. Brook and E. Ben Dor
  • 1Remote Sensing Laboratory, Center for Spatial Analysis Research (UHCSISR), University of Haifa, Haifa, Israel
  • 2Remote Sensing Laboratory, Tel-Aviv University, Tel Aviv, Israel

Keywords: Unmixing, Feature-extraction, L1/2 nonnegative matrix, Orthogonal matching pursuit, Quantitative target detection, Hyperspectral imagery

Abstract. A novel approach for radiometric calibration and atmospheric correction of airborne hyperspectral (HRS) data, termed supervised vicarious calibration (SVC) was proposed by Brook and Ben-Dor in 2010. The present study was aimed at validating this SVC approach by simultaneously using several different airborne HSR sensors that acquired HSR data over several selected sites at the same time. The general goal of this study was to apply a cross-calibration approach to examine the capability and stability of the SVC method and to examine its validity. This paper reports the result of the multi sensors campaign took place over Salon de Provenance, France on behalf of the ValCalHyp project took place in 2011. The SVC method enabled the rectification of the radiometric drift of each sensor and improves their performance significantly. The flight direction of the SVC targets was found to be a critical issue for such correction and recommendations have been set for future utilization of this novel method. The results of the SVC method were examined by comparing ground-truth spectra of several selected validation targets with the image spectra as well as by comparing the classified water quality images generated from all sensors over selected water bodies.