BIN OBJECT RECOGNITION USING IMAGE MATRIX DECOMPOSITION AND NEURAL NETWORKS

Abstract

Bin picking robots require vision sensors capable of recognizing objects in the binirrespective of the orientation and pose of the objects inside the bin. Bin picking systems arestill a challenge to the robot vision research community due to the complexity of segmenting ofoccluded industrial objects as well as recognizing the segmented objects which have irregularshapes. The problem becomes more complex when these objects look like entirely differentobjects in various orientations. In this paper a simple object recognition method is presentedusing singular value decomposition of the object image matrix and a functional link neuralnetwork for a bin picking vision system. The results of the functional link net are comparedwith that of a simple feed forward net. The network is trained using the error back propagationprocedure. The proposed method is robust for recognition of objects.