Correction Algorithm Of Phase Shifting Errors In Phase Measuring Profilometry

Optical 3-D profilometry has been widely used for 3-D sensing, machine vision, intelligent robots control, industry monitoring, biomedicine, dressmaking, etc. Several 3-D object profilometry methods, including Moire technique(MT),phase-measuring profilometry(PMP), Fourier transformation profilometry(FTP), modulation measurement profilometry(MMP) have been exhaustively studied. In these methods, phase-measuring profilometry is a important method because of its higher accuracy. Phase-shifting errors have a great influence on the measuring-accuracy. A few of algorithms have been developed to solve linear phase -shifting errors, but can't solve the problem of random phase-shifting errors. An iterative least-squares fitting method has been proposed by In-Bok Kang to solve random phase shift errors in each interferogram . This method assumes that each inteferogram has an identical phase shift that can be found out by least-suqares fitting. This algorithm determines any errors of phase shifts deviating from their intended values and also suppresses the errors by substituting actual values in subsequent computation .The algorithm can solve the problem of random phase-shifting very accurately, but the algorithm needs a great of iteration number and considerable calculation. A new random phase-shifting correction algorithm ispresented in this paper on the basis of five-step phase-shifting. It can directly solve the value of phase-shifting errors without solving phase. Meanwhile, the new algorithm can greatly decrease the iteration number and calculation without depressing the measurement accuracy. If the phase shifting-errors are enough small (less than 5 degree). It just needs one iteration to solve the values of phase shifts of each inteferogram. In the case of phase shifting-errors being large, it also can take a few of iteration to enhance measurement accuracy. The formula of the new algorithm to solve random phase shift is induced in this paper in detail .The validity of this algorithm has been demonstrated by means of compute simulation. As a result, it can greatly reduce the number of iteration and calculation in comparison with the algorithm of In-Bok Kong.