PSNet: a deep learning model based digital phase shifting algorithm from a single fringe image
This work addresses the problem of limited applicability in optical interferometry and fringe projection profilometry by enabling phase retrieval from a single image, though it is incremental as it builds on existing deep learning methods.
The authors tackled the limitation of phase-shifting algorithms requiring multiple fringe images by proposing PSNet, a deep learning model that predicts additional phase-shift patterns from a single fringe image, achieving promising performance in simulations and experiments with robustness to complex surfaces.
As the gold standard for phase retrieval, phase-shifting algorithm (PS) has been widely used in optical interferometry, fringe projection profilometry, etc. However, capturing multiple fringe patterns in PS limits the algorithm to only a narrow range of application. To this end, a deep learning (DL) model based digital PS algorithm from only a single fringe image is proposed. By training on a simulated dataset of PS fringe patterns, the learnt model, denoted PSNet, can predict fringe patterns with other PS steps when given a pattern with the first PS step. Simulation and experiment results demonstrate the PSNet's promising performance on accurate prediction of digital PS patterns, and robustness to complex scenarios such as surfaces with varying curvature and reflectance.