Evaluation method for noise-induced phase error in fringe projection profilometry

Abstract

For noise-induced phase error and phase unwrapping reliability in fringe projection profilometry (FPP), a new evaluation method is proposed. By introducing a Gaussian noise model, the maximum variance of the noise-induced phase error and phase unwrapping reliability are directly obtained. According to the proposed evaluation method, the variance of the noise-induced phase error is proportional to the noise variance and inversely proportional to the fringe modulation. The phase unwrapping reliability is not only inversely proportional to the noise-induced phase error, but also inversely proportional to the fringe frequency ratio. This work analyzes and compares three efficient dual-frequency phase unwrapping algorithms. The results show that the wrapped phase accuracy and phase unwrapping reliability based on 4 f H + 4 f L are the best, followed by 3 f H + 3 f L and 3 f H + 2 f L . However, 3 f H + 2 f L has the highest measurement efficiency.