Deep DIH: Single-Shot Digital In-Line Holography Reconstruction by Deep Learning

Huayu Li,Abolfazl Razi,Christopher Mann,Zaoyi Chi,Xiwen Chen

doi:10.1109/access.2020.3036380

Huayu Li, Abolfazl Razi + Show 3 more

Open Access

https://doi.org/10.1109/access.2020.3036380

Copy DOI

Abstract

Digital in-line holography is commonly used to reconstruct 3D images from 2D holograms for microscopic objects. One of the technical challenges that arise in the signal processing stage is removing the twin image that is caused by the phase-conjugate wavefront from the recorded holograms. Twin image removal is typically formulated as a non-linear inverse problem due to the irreversible scattering process when generating the hologram. Recently, end-to-end deep learning-based methods have been utilized to reconstruct the object wavefront (as a surrogate for the 3D structure of the object) directly from a single-shot in-line digital hologram. However, massive data pairs are required to train deep learning models for acceptable reconstruction precision. In contrast to typical image processing problems, well-curated datasets for in-line digital holography does not exist. Also, the trained model highly influenced by the morphological properties of the object and hence can vary for different applications. Therefore, data collection can be prohibitively cumbersome in practice as a major hindrance to using deep learning for digital holography. In this paper, we proposed a novel implementation of autoencoder-based deep learning architecture for single-shot hologram reconstruction solely based on the current sample without the need for massive datasets to train the model. The simulations results demonstrate the superior performance of the proposed method compared to the state of the art single-shot compressive digital in-line hologram reconstruction method.

Highlights

Digital holography (DH) is a powerful imaging technique used to reconstruct the three-dimensional (3D) surface of an object from its two dimensional (2D) image captured by a visual sensor
A deep learning method for single-shot reconstruction of in-line digital holography reconstruction is proposed in this paper
The key advantage of the proposed method compared to similar numerical Reconstruction methods and the recently developed deep learning (DL)-based methods is two-fold

Summary

INTRODUCTION

Digital holography (DH) is a powerful imaging technique used to reconstruct the three-dimensional (3D) surface of an object from its two dimensional (2D) image captured by a visual sensor. The captured hologram includes the object field O(x, y) and its conjugation O(x, y), respectively, representing the virtual and real images [6] This phenomenon leads to the formation of the twin image, an issue to be dealt with in the reconstruction phase. It is not directly applicable to digital holography, where the real and imaginary parts of the hologram follow the aforementioned specific wave equations (1-3), and noting the fact that only the intensity of the holographic images is available in our method. Like other uses of autoencoders, no training dataset is required, as a key advantage for our method This method simultaneously performs noise reduction and twin image removal by a well-defined objective function. Experimental results prove the feasibility and the superior performance of the proposed method over the existing CS methods

DEEP LEARNING SCHEME

EXPERIMENTS

SIMULATION RESULTS

Findings

CONCLUSION