Information Based Services: Support to the Analysis of MERIS Data
Michele Iapaolo(1) and Sergio D'Elia(1)
(1) ESA/ESRIN, Via Galileo Galilei, 00044 Frascati, Italy
In recent years, our capability to acquire and archive large volumes of Earth Observation (EO) satellite data has greatly exceeded our capability to extract and timely deliver limited but relevant information from them, as needed by users, service industry and emerging big applications (e.g. global monitoring). The extraction of information is generally time consuming and expensive, and can be therefore systematically applied only to a small subset of the acquired scenes, thus limiting the full exploitation of the huge amount of acquired data.
In many cases, the exploitation could be improved even just by the identification of the potentially relevant images within those available in very large datasets. In this field, Image Information Mining (IIM) techniques are very promising, allowing the automatic identification of relevant images and the support to (or partial replacement of) further processing aiming at information extraction.
The Knowledge-based Information Mining (KIM) tool developed by ESA (in cooperation with DLR, ETHZ and ACS) permits to interactively search large or well characterised areas in all images of a collection, through the combination of automatic primitive feature extraction and machine learning from user interaction. KIM, which is based on a probabilistic approach, was also integrated with the Service Support Environment (SSE)..
This contribution describes the provision of demonstration services through the combined use of KIM and SSE. KIM was trained to probabilistically detect clouds in the MERIS Reduced Resolution (RR) images, systematically ingested from the ESA Rolling Archives. The extracted cloud cover information is stored in the internal database and is provided to the central catalogue of the ESA ground segment. Through this information, the users of dedicated SSE services can identify and download from rolling or permanent archives the MERIS RR images, which have cloud coverage below a user defined threshold within the area of interest.
The work aims at spreading and enlarging the visibility of such Information Based Services within the MERIS community, in order to test them and get useful feedback, also for planning future activities.
Keywords: ESA European
Space Agency - Agence spatiale europeenne,
observation de la terre, earth observation,
satellite remote sensing,
teledetection, geophysique, altimetrie, radar,
chimique atmospherique, geophysics, altimetry, radar,