Phase Retrieval Algorithm Based on Transport of Intensity Equation under the Fusion of Regularization and Grating Modulation

In the quantitative phase retrieval method based on transport of intensity equation, due to the significant amplification of low spatial frequency information caused by the Poisson reversal process, there is a problem of amplifying low-frequency noise, which leads to blurring of the final low-frequency phase results.In this paper, a new approach to the transport of intensity equation phase retrieval method which uses the fusion of regularization and grating modulation is demonstrated. First, we need to aquire two phase images, one is solved by transport of intensity equation under Tikhonov regularization, the other is solved by transport of intensity equation under grating modulation. Then, by solving the minimized power spectral density error, we get the transfer function that can gather the advantages of two retrieval methods. In the end, we combine the two phases using the transfer function to obtain the high-precision hybrid phase retrieval.