Christine Largouët and Marie-Odile Cordier Modélisation par automate temporisé pour aider à l'identification de l'occupation du sol , RFIA'2000 : Reconnaissance des Formes et Intelligence Artificielle , Paris, France , Vol. II , 285-294 , 1-3 Février , 2000 , Document Abstract Landcover recognition from remotely sensed images is often imperfect and leads to ambiguities between crops. In this field, we present an original approach which consists in representing the plot as a dynamic system and in modelling his evolution (knowledge about crop cycles) with a timed automata. Observations, issued from a preclassification of images, are treated with the timed automata, in a prediction/postdiction step to improve results obtained by traditional classification tools. This web site is maintained by René Quiniou using the Weave system from the Caravel project: http://www-caravel.inria.fr. Last modification: 10-07-2011 09:54:25 Version française