Timed Automata Model to Improve the Classification of a Sequence of Images

The aim of this paper is to propose the use of a dynamic plot model to improve landcover classification on a sequence of images. This new approach consists in representing the plot as a dynamic system and in modeling its evolution (knowledge about crop cycles) using the timed automata formalism. In order to refine results obtained by a traditional classifier, observations given by a preliminary classification of images are matched with expected states provided by an automaton simulation. The paper presents the modeling captured by the timed automata formalism and the general method, which is based on prediction and postdiction mechanisms, that have been adopted to improve the classification of a sequence of images. Finally, the interest of the method is demonstrated through experimental results.