Padró J.-C., Muñoz F.-J., Ávila L.Á., Pesquer L., Pons X. (2018) Radiometric correction of Landsat-8 and Sentinel-2A scenes using drone imagery in synergy with field spectroradiometry. Remote Sensing. 10: 0-0.LinkDoi: 10.3390/rs10111687
The main objective of this research is to apply unmanned aerial system (UAS) data in synergy with field spectroradiometry for the accurate radiometric correction of Landsat-8 (L8) and Sentinel-2 (S2) imagery. The central hypothesis is that imagery acquired with multispectral UAS sensors that are well calibrated with highly accurate field measurements can fill in the scale gap between satellite imagery and conventional in situ measurements; this can be possible by sampling a larger area, including difficult-to-access land covers, in less time while simultaneously providing good radiometric quality. With this aim and by using near-coincident L8 and S2 imagery, we applied an upscaling workflow, whereby: (a) UAS-acquired multispectral data was empirically fitted to the reflectance of field measurements, with an extensive set of radiometric references distributed across the spectral domain; (b) drone data was resampled to satellite grids for comparison with the radiometrically corrected L8 and S2 official products (6S-LaSRC and Sen2Cor-SNAP, respectively) and the CorRad-MiraMon algorithm using pseudo-invariant areas, such as reflectance references (PIA-MiraMon), to examine their overall accuracy; (c) then, a subset of UAS data was used as reflectance references, in combination with the CorRad-MiraMon algorithm (UAS-MiraMon), to radiometrically correct the matching bands of UAS, L8, and S2; and (d) radiometrically corrected L8 and S2 scenes obtained with UAS-MiraMon were intercompared (intersensor coherence). In the first upscaling step, the results showed a good correlation between the field spectroradiometric measurements and the drone data in all evaluated bands (R2 > 0.946). In the second upscaling step, drone data indicated good agreement (estimated from root mean square error, RMSE) with the satellite official products in visible (VIS) bands (RMSEVIS < 2.484%), but yielded poor results in the near-infrared (NIR) band (RMSENIR > 6.688% was not very good due to spectral sensor response differences). In the third step, UAS-MiraMon indicated better agreement (RMSEVIS < 2.018%) than the other satellite radiometric correction methods in visible bands (6S-LaSRC (RMSE < 2.680%), Sen2Cor-SNAP (RMSE < 2.192%), and PIA-MiraMon (RMSE < 3.130%), but did not achieve sufficient results in the NIR band (RMSENIR < 7.530%); this also occurred with all other methods. In the intercomparison step, the UAS-MiraMon method achieved an excellent intersensor (L8-S2) coherence (RMSEVIS < 1%). The UAS-sampled area involved 51 L8 (30 m) pixels, 143 S2 (20 m) pixels, and 517 S2 (10 m) pixels. The drone time needed to cover this area was only 10 min, including areas that were difficult to access. The systematic sampling of the study area was achieved with a pixel size of 6 cm, and the raster nature of the sampling allowed for an easy but rigorous resampling of UAS data to the different satellite grids. These advances improve human capacities for conventional field spectroradiometry samplings. However, our study also shows that field spectroradiometry is the backbone that supports the full upscaling workflow. In conclusion, the synergy between field spectroradiometry, UAS sensors, and Landsat-like satellite data can be a useful tool for accurate radiometric corrections used in local environmental studies or the monitoring of protected areas around the world. © 2018 by the authors.
Pons X., Pesquer L., Cristobal J., Gonzalez-Guerrero O. (2014) Automatic and improved radiometric correction of landsat imageryusing reference values from MODIS surface reflectance images. International Journal of Applied Earth Observation and Geoinformation. 33: 243-254.LinkDoi: 10.1016/j.jag.2014.06.002
Radiometric correction is a prerequisite for generating high-quality scientific data, making it possibleto discriminate between product artefacts and real changes in Earth processes as well as accuratelyproduce land cover maps and detect changes. This work contributes to the automatic generation of surfacereflectance products for Landsat satellite series. Surface reflectances are generated by a new approachdeveloped from a previous simplified radiometric (atmospheric + topographic) correction model. Theproposed model keeps the core of the old model (incidence angles and cast-shadows through a digitalelevation model [DEM], Earth-Sun distance, etc.) and adds new characteristics to enhance and automatizeground reflectance retrieval. The new model includes the following new features: (1) A fitting model basedon reference values from pseudoinvariant areas that have been automatically extracted from existingreflectance products (Terra MODIS MOD09GA) that were selected also automatically by applying qualitycriteria that include a geostatistical pattern model. This guarantees the consistency of the internal andexternal series, making it unnecessary to provide extra atmospheric data for the acquisition date and time,dark objects or dense vegetation. (2) A spatial model for atmospheric optical depth that uses detailedDEM and MODTRAN simulations. (3) It is designed so that large time-series of images can be processedautomatically to produce consistent Landsat surface reflectance time-series. (4) The approach can handlemost images, acquired now or in the past, regardless of the processing system, with the exception ofthose with extremely high cloud coverage. The new methodology has been successfully applied to aseries of near 300 images of the same area including MSS, TM and ETM+ imagery as well as to differentformats and processing systems (LPGS and NLAPS from the USGS; CEOS from ESA) for different degreesof cloud coverage (up to 60%) and SLC-off. Reflectance products have been validated with some exampleapplications: time series robustness (for a pixel in a pseudoinvariant area, deviations are only 1.04% onaverage along the series), spectral signatures generation (visually coherent with the MODIS ones, butmore similar between dates), and classification (up to 4 percent points better than those obtained withthe original manual method or the CDR products). In conclusion, this new approach, that could also beapplied to other sensors with similar band configurations, offers a fully automatic and reasonably goodprocedure for the new era of long time-series of spatially detailed global remote sensing data. © 2014 The Authors.
Pesquer L., Domingo C., Pons X. (2013) A geostatistical approach for selecting the highest quality MODIS daily images. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). 7887 LNCS: 608-615.LinkDoi: 10.1007/978-3-642-38628-2_72
The aim of this work was to develop a new methodology for automatic selection of the highest quality MODIS daily images, MOD09GA Surface Reflectance product. The methodology developed here complements the quality assessment of MODIS products with a geostatistical analysis of spatial pattern images based on variogram tools. The resulting selection is formed by 26 high-quality images (from an initial dataset of 365) from throughout 2007. Most images with geometric distortion problems, such as the bow-tie effect, were rejected. The automatic selection was validated by comparing it to manual selection, which showed that it achieved an overall accuracy of 71.4%. © 2013 Springer-Verlag.
Pesquer L., Pons X., Cortes A., Serral I. (2013) Spatial pattern alterations from JPEG2000 lossy compression of remote sensing images: Massive variogram analysis in high performance computing. Journal of Applied Remote Sensing. 7: 0-0.LinkDoi: 10.1117/1.JRS.7.073595
We evaluate the implications of JPEG2000 lossy compression of remote sensing images for spatial analytical purposes. The main issue is to identify which cases and conditions in geostatistical studies are suitable for using lossy compressed images. For these purposes, an extensive test using Landsat, compact airborne spectrographic imager (CASI), and moderate resolution imaging spectroradiometer (MODIS) image series has been analyzed, through applying and comparing two-dimensional and three-dimensional (spectral and time domains) compression methods with a wide range of compression ratios for several dates, different landscape regions, and spectral bands. Due to the massive test bed and consequently to the high time consuming executions, a parallel solution was specifically developed. Variogram analyses showed that all the compression ratios maintain the variogram shapes, but high compression ratios (>20:1) degrade the spatial patterns of the remote sensing images. These alterations are lower for the three-dimensional compression method, which was a considerable improvement (25%) on the two-dimensional method for large three-dimensional series (CASI, MODIS). However, the two methods behave similarly in the Landsat case. Finally, the parallel solution in a distributed environment demonstrates that high performance computing offers a suitable scientific platform for highly demanding time execution applications, such as geostatistical analyses of remote sensing images. © The Authors. Published by SPIE under a Creative Commons Attribution 3.0 Unported License.
Subscribe to our Newsletter to get the lastest CREAF news.
© 2016 CREAF | Legal notice