Calibration of 3D surface profilometry using digital fringe projection

An effective calibration method, by minimizing measurement errors, has been developed to increase the accuracy of 3D profilometry using digital fringe projection and phase-shifting method. In digital fringe projection, the image intensity and distribution of the sinusoidal fringe patterns projected on the measured surface can be critically affected by lens distortions and image aberrations. The phase difference calculated by the phase-shift principle can be significantly influenced by these error sources and become nonlinear to the optical phase difference (OPD) existing between the surface profiles. This paper demonstrates a 3D calibration method developed to obtain accurate system parameters for 3D surface measurement. The calibration method utilizes a known accurate 3D calibrating block and projection mathematical models for identification of the system parameters by means of least-squares minimization. Accurate clouds of 3D data points can be obtained by a 3D mapping method between the object space and the image coordinates incorporating the phase difference. The measurement accuracy of surface contouring can be maintained well within 2% of the overall measurement range. Verified with the experimental results, the proposed calibration method can effectively reduce more than 60% of the maximum measured error in comparison with the traditional phase-conversion method.