Optical recognition of three-dimensional objects with in-plane rotation invariance

Abstract

In this paper we present a real-time system for three- dimensional (3D) object recognition. It is based on the use of the Fourier transform profilometry technique and a joint transform correlator. The projection of a regular fringe pattern onto the 3D object surface permits us to obtain a distorted grating pattern that carries all the 3D information of the object. The analysis of such patterns is the basis of the method for recognizing 3D objects in real time. Later, by encoding with a fringe pattern a single component of the circular harmonic expansion derived from the object to be detected, a rotation-invariant 3D object recognition is also demonstrated. The whole experimental setup is robust and can be constructed with simple equipment. The experimental results demonstrate the theory and show the utility of the proposed method.