A robust and effective phase-shift fringe projection profilometry method for the extreme intensity

摘要

The structured-light with phase-shift fringe projection profilometry (PFPP) plays a significant role in non-contact 3D surface measurement. The multi-frequency heterodyne (MFH) phase unwrapping algorithm is widely used in PFPP because of its high precision and noise-isolation features. However, measuring the over-dark and over-bright surfaces is a headache of the MFH algorithm due to its limit intensity tolerance in decoding phases. We provide a new method named as Moire Fringe Patterns Sequence (MFPS) algorithm to avoid the phase error enlargement, so as to increase the robustness of the decoding algorithm. In the MFPS algorithm, we project the synthetic patterns to the object directly, which doubles the tolerance of phase error. The MFPS algorithm has the same accuracy as the MFH. It can successfully decode the phase of the extreme pixel intensity (as low as 10 or saturated at 255), but require no more hardware or additional work. The theoretical, simulation and experimental studies verify that the MFPS algorithm is robust and effective in measuring the over-dark and over-bright object.