The header file UrbanSpectralLibrary_APEX_Brussels_2015_GENLIB.hdr and its associated spectral library file UrbanSpectralLibrary_APEX_Brussels_2015_GENLIB.sli constitute a library of 1274 urban cover type spectra describing surface reflectance over the VNIR-SWIR wavelength domain. The library spectra were extracted from a 2m-resolution APEX airborne hyperspectral image of Brussels (SONIA site of BELAIR program) acquired during the summer of 2015. The source image can be downloaded from the VITO Product Distribution Portal. Endmember pixels were algorithmically selected using Iterative Spectral Distancing (see this paper for more info), and the corresponding endmember spectra were smoothed with a Gaussian filter. Note that the WGS84 latitude-longitude coordinates of the endmember pixels are included in the header file. The spectra were manually interpreted and labelled using high-resolution aerial imagery, Google Street View, online documentation and comparison to previously published urban spectral libraries (SLUM, KLUM, UCSB ...). Three sets of spectra labels are provided in this library: one for EAGLE land cover (EAGLE_LCC, consult the EAGLE website for more info), one for generalized material groupings (GENLIB_LCH_BuC_MG) and one for more detailed artificial material type (GENLIB_LCH_BuC_AMT).

If you use this data set for published work, please cite it correctly. Suggested citation format: "Priem, F., Somers, B., Canters, F. (2022). Brussels image-based library of urban materials, version X [Data set]. https://doi.org/10.5281/zenodo.6472398". This spectral library was produced in the frame of the GENLIB research project that received funding from the Belgian Science Policy Office per the agreement of subsidy number SR/67/167 (BELAIR 2013), SR/XX/170 (BELAIR SONIA), SR/03/333 (BELAIR SONIA 2015) and SR/00/378 (GENLIB). While every effort was made to ensure accurate information, this data set is presented "as is" without warranties of any kind. The authors accepts no liability or responsibility to any person as a consequence of any reliance upon the data presented here. The user assumes all responsibility and risk for the use of this data. © Priem 2022