Model-Based Objects Recognition in Man-Made Environments

In this paper we describe a model-based objects recognition system which is par t of an image interpretation system intended to assist autonomous vehtcles navigation. The system is intended to operate in manmade environments. Behavior-based navigation of autonomous vehicles involves the recognition of navagable areas and the potential obstacles. The recognition system integrates color, shape and texture information together with the location of the vanishing point. The recognitzon process starts from some prior scene knowledge, that is, a generic model of the expected scene and the potenczal obiects. The recognition system constttutes an approach where different low-level vision techniques extract a multitude of image descriptors whtch are ihen analyzed using a rule-based reasoning system to tnterpret the image content. This system has been tmplemented using CEES, the C++ Embedded Expert System Shell developed in the Systems Engineenng and Automatzc Control Laboratory (Untverszty of Girona) as a specific rule-based problem soluzng tool. It has been especially conetrtved for supporting cooperaiive ETped $piem$, and uses the object onented programmzng paradigm.