Phase retrieval for Fourier Ptychography under varying amount of measurements
This work addresses acquisition time and data challenges in high-resolution imaging for optical systems, though it appears incremental as it builds on existing deep learning approaches.
The paper tackles the phase retrieval problem in Fourier Ptychography, where traditional methods fail under low overlap conditions, by proposing an auto-encoder based architecture that adaptively trains for both low and high overlap cases, outperforming previous techniques in simulations with uncorrelated phase and amplitude.
Fourier Ptychography is a recently proposed imaging technique that yields high-resolution images by computationally transcending the diffraction blur of an optical system. At the crux of this method is the phase retrieval algorithm, which is used for computationally stitching together low-resolution images taken under varying illumination angles of a coherent light source. However, the traditional iterative phase retrieval technique relies heavily on the initialization and also need a good amount of overlap in the Fourier domain for the successively captured low-resolution images, thus increasing the acquisition time and data. We show that an auto-encoder based architecture can be adaptively trained for phase retrieval under both low overlap, where traditional techniques completely fail, and at higher levels of overlap. For the low overlap case we show that a supervised deep learning technique using an autoencoder generator is a good choice for solving the Fourier ptychography problem. And for the high overlap case, we show that optimizing the generator for reducing the forward model error is an appropriate choice. Using simulations for the challenging case of uncorrelated phase and amplitude, we show that our method outperforms many of the previously proposed Fourier ptychography phase retrieval techniques.