Dynamic threshold cloud detection algorithm improvement for AVHRR and SEVIRI images

The aim of this work is to present a simple and fast automatic cloud detection algorithm for Advanced Very High Resolution Radiometer (AVHRR) and Spinning Enhanced Visible and Infrared Imager (SEVIRI) satellite images. The algorithm was developed by the Department of Electronics and Telecommunications (University of Florence) for the Satellite Receiving Station of Prato Campus (University of Florence), where AVHRR and SEVIRI data have been directly received since 1997. The algorithm is designed to meet the need for real-time operational processing of land and sea products, such as vegetation indexes and regional land/sea surface temperature maps (i.e. Italy). It is developed as simple and fast processing which does not need to use ancillary data. The algorithm is tested for AVHRR and SEVIRI images directly received at the Station which are characterized by different percentages of cloudy pixels. Algorithm results are compared with control cloud masks, which are created manually by a visual inspection of the image to be cloud screened.