Title:
Fourier ptychographic imaging via neural network modeling
Abstract(s):
Fourier ptychography is a recently developed phase retrieval approach for large field-of-view and high-resolution microscopy. This technique stitches together many variably illuminated, low-resolution measurements in the Fourier space to expand the frequency passband and recover the high-resolution complex sample image. Without involving any mechanical scanning, it facilitates large field of view imaging in a simple and robust manner. In this talk, I will discuss the principle of the Fourier ptychography approach and its applications. I will also discuss how to model the Fourier ptychographic forward imaging process using a neural network and recover the complex object information in a network training process. In the proposed neural network, the object is treated as 2D learnable weights of a convolution or a multiplication layer. The output of the network is modeled as the loss function one aims to minimize. The batch size of the network corresponds to the number of captured low-resolution images in one forward / backward pass. Since convolution and multiplication are the two most-common operations in imaging modeling, the proposed approach may provide a new perspective to examine many coherent and incoherent systems.
Speaker Bio(s):
Dr. Guoan Zheng received the M.S. and Ph.D. degrees in electrical engineering from Caltech in 2008 and 2013, respectively. He is currently an assistant professor at the University of Connecticut. He is the recipient of UConn research excellence award, Caltech Demetriades Thesis Prize, and Lemelson-MIT Caltech Student Prize. His research interest focuses on the development of novel imaging and sensing techniques for biomedical applications.