Somafid: a Seco D Level Stratified Image Processi G System for Thermal a Omalies Ide Tificatio I (a)atsr Imagery

Abstract

The detection of surface thermal anomalies and the subsequent identification of fire pixels is a central issue for geostationary satellites and medium resolution polar orbit satellites. Presently a limited number of sensors can be used to detect thermal anomalies from space; among them, the (A)ATSR (Advanced Along Track Scanning Radiometer) sensor series has been used for years for this scope. (A)ATSR sensors have adequate spectral characteristics to detect ground thermal anomalies, but the low saturation level of the 3.7 micron channel drastically reduces their applicability over bright regions. During last years MEEO has developed a MODIS fire detection system named SOMAFID (SOIL MAPPER® FIre Detection system); SOMAFID is a Thermal Anomalies / Fire Detection system based on the original algorithm of the MOD14 thermal anomalies product [5]. The use of the SOIL MAPPER® preliminary classification system [9] as a preliminary map permits to improve the MOD14 performance. With respect to the MOD14 product, the SOMAFID system reduces false alarms (bright object mask improved) and improves the quality of the final product introducing three background types over which fire pixels are detected; SOMAFID generates nine classes: three types of fire pixels (Flaming, Smoldering, Mixed, by means of the Kaufman model [6]), over three different background types (high vegetation, low vegetation, no vegetation). A version of the SOMAFID software has been implemented also for (A)ATSR: the low saturation level of the 3.7 micron of the (A)ATSR sensor series do not allow to discriminate the nature of fire pixels (if a fire pixel is a flaming fire pixel, a smoldering fire pixel or a mix of both). Currently, the (A)ATSR version of SOMAFID is adopted in the framework of the Classification Application-services and Reference Datasets (CARD) project [11]: a web service on the Service Support Environment (SSE) ESA Portal is available to the community free of charge until June 3

9 Figures and Tables

Cite this paper

@inproceedings{Pellegrini2008SomafidAS, title={Somafid: a Seco D Level Stratified Image Processi G System for Thermal a Omalies Ide Tificatio I (a)atsr Imagery}, author={L. Pellegrini and Stefano atali and Andrea Baraldi}, year={2008} }