The use of very high resolution (VHR) aerial imagery for quantitative remote sensing has been limited by unwanted radiometric variation over temporal and spatial extents. In this paper we propose a simple yet effective technique for the radiometric homogenisation of the digital numbers
of aerial images. The technique requires a collocated and concurrent, well-calibrated satellite image as surface reflectance reference to which the aerial images are calibrated. The bands of the reference satellite sensor should be spectrally similar to those of the aerial sensor. Using radiative
transfer theory, we show that a spatially varying local linear model can be used to approximate the relationship between the surface reflectance of the reference image and the digital numbers of the aerial images. The model parameters for each satellite pixel location are estimated using least
squares regression inside a small sliding window. The technique was applied to a set of aerial images captured over multiple days with an Intergraph Digital Mapping Camera (DMC) system. A near-concurrent Moderate Resolution Imaging Spectroradiometer (MODIS) nadir bidirectional reflectance
distribution function (BRDF) adjusted reflectance image was used as the reflectance reference dataset. The resulting DMC mosaic was compared to a near-concurrent Satellite Pour l’Observation de la Terre (SPOT) 5 reflectance image of a portion of the same area, omitting the blue channel
from the DMC mosaic due to its absence in the SPOT 5 data. The mean absolute reflectance difference was found to be 3.43% and the mean coefficient of determination (R
2) over the bands was 0.84. The technique allows the production of seamless mosaics corrected for coarse scale
atmospheric and BRDF effects and does not require the manual acquisition (or provision) of ground reflectance references. The accuracy of corrections is limited by the resolution of the reference image, which is generally significantly coarser than VHR imagery. The method cannot correct for
small scale BRDF or other variations not captured at the reference resolution. Nevertheless, results show a significant improvement in homogeneity and correlation with SPOT 5 reflectance.
No Reference information available - sign in for access.
No Citation information available - sign in for access.
No Supplementary Data.
No Article Media
Document Type: Research Article
Department of Geography and Environmental Studies, Stellenbosch University, Stellenbosch, South Africa
Centre for Geographical Analysis, Stellenbosch University, Stellenbosch, South Africa
Publication date: April 3, 2019
More about this publication?