Model Based Recognition Of 2-D Occluded Objects

A model based method for the recognition of 2-D occluded objects is presented. Initially, to reduce the number of models to be considered for identification from the model database, signatures comprising 5-tuple curvatures are used. Next, point to point matching known as exact matching is used to find the correct models present in the scene. The results are illustrated for an occluded scene consisting of industrial objects.