Super typhoon Haiyan makes landfall
Super typhoon Haiyan
Subscribe here to receive special images


Lake Maggiore - Italy, 3-6 April 2000

Detection of Clouds and Cloud-Shadows for VEGETATION images


Paper (pdf file, 92 k)

Cloud detection is an essential part of the pre-processing chain at the CTIV (Centre de Traitement d’Images Vegetation) for various products of the VEGETATION sensor. State of the art techniques have been developed to construct a 3 level cloud mask and a binary cloud shadow mask. Several approaches to the cloud masking problem can be found in literature on other sensors (AVHRR, MODIS), but the challenge in case of the VEGETATION sensor is the absence of channels in the thermal infrared section of the spectrum. This severely complicates cloud-masking efforts, because the thermal information is of crucial importance: clouds in general are far colder than the earth's surface. A genetic algorithm has been developed to optimize thresholds on the VEGETATION spectral bands. The increase in performance is dramatic, and improves the quality of VEGETATION products (atmospheric correction, synthesis…).

Clouds also cast shadows on the earth's surface. This can lead up to a 40% bias of the true reflectance of the underlying terrain element. A new technique has been developed to provide a cloud shadow mask. The position of shadows can be theoretically computed using geometry. Sun and viewing angles are known but cloud height is not. The innovating aspect of the proposed method is to predict cloud height, by measuring the distance from cloud to shadow edge (detected from differences in radiometry). Shadows then, are calculated by means of geometry. This differs from existing techniques that predominantly rely on radiometry and are therefore very sensitive to the reflectivity of the surface. The results for the cloud shadow mask are very promising.