CLOUD AND CLOUD SHADOW MASKING OF HIGH AND MEDIUM RESOLUTION OPTICAL SENSORS – AN ALGORITHM INTER-COMPARISON EXAMPLE FOR LANDSAT 8

摘要

Image processing for satellite water quality products requires reliable cloud and cloud shadow detection and cloud classification before atmospheric correction. Within the FP7/HIGHROC (“HIGH spatial and temporal Resolution Ocean Colour”) Project, it was necessary to improve cloud detection and the cloud classification algorithms for the spatial high resolution sensors, aiming at Sentinel 2 and using Landsat 8 as a precursor. We present a comparison of three different algorithms, AFAR developed by RBINS; ACCAm created by VITO, and IDEPIX developed by Brockmann Consult. We show image comparisons and the results of the comparison using a pixel identification database (PixBox); FMASK results are also presented as reference.