Abstract
Image reconstruction using minimal measured information has been a long-standing open problem in many computational imaging approaches, in particular in-line holography. Many solutions are devised based on compressive sensing (CS) techniques with handcrafted image priors or supervised deep neural networks (DNN). However, the limited performance of CS methods due to lack of information about the image priors and the requirement of an enormous amount of per-sample-type training resources for DNNs has posed new challenges over the primary problem. In this study, we propose a single-shot lensless in-line holographic reconstruction method using an untrained deep neural network which is incorporated with a physical image formation algorithm. We demonstrate that by modifying a deep decoder network with simple regularizers, a Gabor hologram can be inversely reconstructed via a minimization process that is constrained by a deep image prior. The outcoming model allows to accurately recover the phase and amplitude images without any training dataset, excess measurements, or specific assumptions about the object’s or the measurement’s characteristics.
Highlights
Image reconstruction using minimal measured information has been a long-standing open problem in many computational imaging approaches, in particular in-line holography
We demonstrate that by utilizing simple and commonly available regularization methods, a decoder network (DDN) with a physical image formation algorithm can be upgraded to a powerful compressive signal reconstruction model
Due to the small sample-to-sensor distance of the adopted line Holographic Microscopy (LIHM) experimental setup, the field information is encoded within intensely entangled interferometric patterns from which, TwIST algorithm cannot properly reconstruct complex geometries
Summary
Image reconstruction using minimal measured information has been a long-standing open problem in many computational imaging approaches, in particular in-line holography. One can record multiple holograms in different sample-to-sensor distances and recover the complete object information through a physics-based iterative p rocess[6,14,15] Such multi-holograms can be retrieved from multiple wavelengths16, angles[17], or phase s hifts[18,19]; generally derived from the methods of alternating projections[11,12,13,20,21]. Despite their robustness, Multiple measurements are the backbone of these approaches that limits the usage of LIHM for different imaging problems, especially with poor image acquisition rate. Different studies have demonstrated the state-of-the-art performance of DNNs in various imaging problems[27,28,29,30,31] such as L IHM32,33
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
