3-D object recognition by matching the total view information

In this paper we present a complete 3-D object recognition system. The system has the following components: (1) The applied, viewer centered extended surface normal image (ESNI) representation is new and promising. (2) Photometric stereo method (PSM)-we have integrated specular highlight elimination and we have developed a method, that combines dense gradient information and a line drawing of the analyzed objects. (3) An adaptive indexing of geometric features and color indexing is used as a preselecting matching module. (4) A main feature of the system is the 3-D rotation invariance of the unique base face orientation achieved by PSM. An intelligent template matching method is introduced to refine the recognition. The method is powerful but it assumes a fixed 3-D orientation of the objects. Because of the achieved (well-defined) base face orientation this restriction is overcome. We show results of experiments with real, deformed and dirty goods packages