Single-shot lensless imaging with fresnel zone aperture and incoherent illumination

Wu, Juanjuan; Zhang, Hua; Zhang, Wenhui; Jin, Guofan; Cao, Liangcai; Barbastathis, George

doi:10.1038/s41377-020-0289-9

Cited by 129 publications

(47 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Figure 3 c depicts the results obtained by TwIST (Two-step Iterative Shrinkage/Thresholding 42 ) algorithm accompanied with a TV-regularizer (Supplementary section A ). This approach is widely used to compressively solve inverse problems as holographic reconstruction 24 , 25 , 43 . Due to the small sample-to-sensor distance of the adopted LIHM experimental setup, the field information is encoded within intensely entangled interferometric patterns from which, TwIST algorithm cannot properly reconstruct complex geometries.…”

Section: Resultsmentioning

confidence: 99%

Holographic optical field recovery using a regularized untrained deep decoder network

Niknam

Qazvini

Latifi

2021

Sci Rep

View full text Add to dashboard Cite

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.

show abstract

Section: Resultsmentioning

confidence: 99%

Holographic optical field recovery using a regularized untrained deep decoder network

Niknam

Qazvini

Latifi

2021

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Dynamic range, sensitivity, and low noise of an image sensor are now particularly important to obtain an incoherent digital hologram because the visibility and light intensity of such a hologram are lower than those of a hologram generated with a laser. Its importance on advanced digital signal processing techniques based on informatics such as deep learning and compressive sensing will also be increased (Wu et al, 2020;Wu et al, 2021). Algorithms for accelerating holographic image reconstruction are continuously developed and these will lead to real-time holographic measurement (Tsuruta et al, 2021;Shimobaba et al, 2022).…”

Section: Discussionmentioning

confidence: 99%

“…Any light source including self-luminous light and natural light can be recorded as a hologram. Various techniques such as 3D microscopy for self-luminous light (Schilling et al, 1997;Rosen and Brooker, 2008;Jang et al, 2015;Yanagawa et al, 2015;Quan et al, 2017;Liebel et al, 2020a;Liebel et al, 2020b;Kumar et al, 2020;Marar and Kner, 2020;Potcoava et al, 2021), 3D imaging with a single image sensor (Kim, 2013;Clark and Kim, 2015;Jang et al, 2015;Vijayakumar et al, 2016;Vijayakumar and Rosen, 2017;Nobukawa et al, 2018;Choi et al, 2019;Nobukawa et al, 2019;Wu et al, 2020;Tahara et al, 2021d;Wu et al, 2021;Yoneda et al, 2021;Tahara et al, 2022), 3D thermal measurement (Imbe, 2019), and spectroscopic 3D imaging (Yoshimori, 2001;Teeranutranont and Yoshimori, 2013;Naik et al, 2014;Kalenkov et al, 2019) have been proposed to date. Research on applications of IDH is ongoing and a challenging theme.…”

Section: Applications Of Incoherent Digital Holographymentioning

confidence: 99%

Review of Incoherent Digital Holography: Applications to Multidimensional Incoherent Digital Holographic Microscopy and Palm-Sized Digital Holographic Recorder—Holosensor

Tahara

2022

Front. Photon.

View full text Add to dashboard Cite

We review advancements in incoherent digital holography (IDH) with an image sensor and its applications to multidimensional microscopy and a palm-sized hologram recorder termed “holosensor”. There are two types of representative IDH technique: IDH with a structured illumination and a single photodetector termed optical scanning holography and self-interference IDH. The latter IDH is a technique to obtain an incoherent digital hologram by modulating an incoherent light wave between an object and an image sensor. Multidimensional information such as three-dimensional space and wavelengths is simultaneously recorded without changing optical filters by introducing interferometric techniques invented in laser holography. Applications to high-speed color-multiplexed holographic fluorescence microscopy, single-shot incoherent full-color holographic microscopy with white light, and a palm-sized multidimensional incoherent hologram recorder have been developed using multidimensional IDH systems. Schematics and experimental results obtained using IDH techniques, incoherent holographic microscopy systems, and compact IDH systems are introduced.

show abstract

“…Here we reproduce the object based on the detector intensity patterns and assume that the detector is in the focal plane associated with a quadratic radial phase of the mask. In recent work, Fresnel mid-field imaging shows potential for better object-based depth detection [42].…”

Section: The Hybrid Vision Systemmentioning

confidence: 99%

Toward simple, generalizable neural networks with universal training for low-SWaP hybrid vision

et al. 2021

View full text Add to dashboard Cite

Speed, generalizability, and robustness are fundamental issues for building lightweight computational cameras. Here we demonstrate generalizable image reconstruction with the simplest of hybrid machine vision systems: linear optical preprocessors combined with nohidden-layer, "small-brain" neural networks. Surprisingly, such simple neural networks are capable of learning the image reconstruction from a range of coded diffraction patterns using two masks. We investigate the possibility of generalized or "universal training" with these small brains. Neural networks trained with sinusoidal or random patterns uniformly distribute errors around a reconstructed image, whereas models trained with a combination of sharp and curved shapes (the phase pattern of optical vortices) reconstruct edges more boldly. We illustrate variable convergence of these simple neural networks and relate learnability of an image to its singular value decomposition entropy (SVD-entropy) of the image. We also provide heuristic experimental results. With thresholding, we achieve robust reconstruction of various disjoint datasets. Our work is favorable for future real-time low-SWaP hybrid vision: we reconstruct images on a 15W laptop CPU with 15k fps: faster by a factor of 3 than previously reported results and 3 orders of magnitude faster than convolutional neural networks.

show abstract

Single-shot lensless imaging with fresnel zone aperture and incoherent illumination

Cited by 129 publications

References 47 publications

Holographic optical field recovery using a regularized untrained deep decoder network

Holographic optical field recovery using a regularized untrained deep decoder network

Review of Incoherent Digital Holography: Applications to Multidimensional Incoherent Digital Holographic Microscopy and Palm-Sized Digital Holographic Recorder—Holosensor

Toward simple, generalizable neural networks with universal training for low-SWaP hybrid vision

Contact Info

Product

Resources

About