Learning-based complex field recovery from digital hologram with various depth objects

Abstract

In this paper, we investigate a learning-based complex field recovery technique of an object from its digital hologram. Most of the previous learning-based approaches first propagate the captured hologram to the object plane and then suppress the DC and conjugate noise in the reconstruction. To the contrary, the proposed technique utilizes a deep learning network to extract the object complex field in the hologram plane directly, making it robust to the object depth variations and well suited for three-dimensional objects. Unlike the previous approaches which concentrate on transparent biological samples having near-uniform amplitude, the proposed technique is applied to more general objects which have large amplitude variations. The proposed technique is verified by numerical simulations and optical experiments, demonstrating its feasibility.

HoloTile light engine: new digital holographic modalities and applications

“…Even for machine learning methods wherein aberration correction is not the explicit goal, the end-to-end design of certain network architectures ensures implicit correction, as long as these corrections are either constant between training and inference, or general enough during training to be useful for correcting arbitrary aberrations. For instance, several strides have been made in ML-aided holographic reconstruction [78][79][80][81][82][83][84][85][86][87] in which either raw holograms or back-propagated holograms are processed in custom trained networks to generate more accurate, aberration-reduced, reconstructions.…”

Generative adversarial neural network for 3D-hologram reconstruction

“…CNNs can be used for hologram generation [20][21][22] and reconstruction, including noise, twin image and zero-order suppression [23]. CNNs are typically used to reconstruct one image [24][25][26][27][28][29][30][31][32] or two (amplitude and phase information) [33][34][35][36][37][38][39][40][41] or extended focus imaging [41,42]. However, the direct reconstruction of the entire 3D-scene provides a wider range of possibilities [43].…”

HoloForkNet: Digital Hologram Reconstruction via Multibranch Neural Network

“…For holographic image reconstruction, the focus is mainly on obtaining a single amplitude or phase [48][49][50] image or two images: amplitude/phase [36,[51][52][53][54][55][56][57] and extended focused imaging/depth map ones [57,58], in a single [36,[51][52][53][54][56][57][58] or multiwavelength [55] scheme. Typically, neural networks for 3D hologram reconstruction contain one decoder branch, reconstructing not object planes separately but a depth map where objects from different planes are separated by their intensity.…”

“…tasets, which led to the development of neural network-based algorith tasks. In particular, in holography, neural network-based methods can be struction acceleration [34], denoising [35][36][37], aberrations compensation [3 sion of parasitic diffraction orders [40,41], data compression [42], particle lo sitioning [43,44] and classification [45], depth prediction and autofocusing For holographic image reconstruction, the focus is mainly on obtainin plitude or phase [48][49][50] image or two images: amplitude/phase [36,[51][52][53][54][55][56][57] focused imaging/depth map ones [57,58], in a single [36,[51][52][53][54][56][57][58] or m [55] scheme. Typically, neural networks for 3D hologram reconstruction c coder branch, reconstructing not object planes separately but a depth map from different planes are separated by their intensity.…”