Parameterized Reconstruction Based Fourier Ptychography

Weixin Jiang,Yongbing Zhang,Qionghai Dai
DOI: https://doi.org/10.1109/icme.2016.7552879
2016-01-01
Abstract:Fourier ptychography (FP) is recently proposed as a computational imaging technique, which aims at enhancing the space-bandwidth product (SBP) of the optical imaging system. Specifically, the FP recovery routine iteratively stitches together a number of low-resolution images, which are captured under angularly varying illumination, to produce a wide-field, high-resolution image. However, the reconstruction procedure of the FP recovery routine is based on a low efficient phase retrieval algorithm and may severely degrade quality of the reconstructions. To address this problem, in this paper, we develop and test a Parameterized Reconstruction based Fourier Ptychography (PR-FP), which parameterizes the reconstruction procedure by introducing an updating-coefficient. Meanwhile, a convergence-related metric, which measures how good the reconstruction matches the input dataset, is proposed to help determine a proper updating-coefficient. Extensive experimental results demonstrate that the proposed PR-FP algorithm achieves superior reconstructions both on simulated dataset and real captured dataset.
What problem does this paper attempt to address?