Multi-scale digital holographic reconstruction with deep learning

Huaying Wang,Qiwen Li,Shuo Wang,Gaofu Men

DOI: https://doi.org/10.1364/ao.541739

IF: 1.9

2024-12-14

Applied Optics

Abstract:Huaying Wang, Qiwen Li, Shuo Wang, Gaofu Men Aimed at the problems of complex operation and zero-order influence in the reconstruction of holograms of multi-scale and large objects, a ... [Appl. Opt. 64, B31-B37 (2025)]

optics

What problem does this paper attempt to address?

Multiplane Digital Holography Based on Extrapolation Iterations

Zhengzhong Huang,Cuifang Kuang,Lin Xu,Liangcai Cao

DOI: https://doi.org/10.1016/j.optcom.2020.126526

IF: 2.4

2020-01-01

Optics Communications

Abstract:The resolution of reconstructed images is significantly limited by the size of captured holograms in digital holography. A multiplane digital holographic algorithm based on extrapolation iterations is proposed herein. Holograms at different axial distances are captured with a shifting imaging sensor and then extrapolated to a larger hologram after padding. Multiple holograms provide multipoint constraints in diffraction calculation. Constraints based on the energy distribution and transmissions of an object are applied to extract the object function from the holograms. The method achieves a higher convergence speed for solving the inverse problem compared with the single-plane extrapolation method. We demonstrate that the reconstruction in multiple planes yields a better resolution than that in one plane.
Deep Learning Achieves Super-resolution in Digital Holography.

Zhi Zhong,Qi Cong,Mingguang Shan

DOI: https://doi.org/10.1145/3408127.3408195

2020-01-01

Abstract:Digital holographic microscopy is a non-contact, full-field quantitative measurement and unlabeled fast 3D measurement method. During the recording process of holograms, high frequency information is lost due to interference from external factors and human factors. In this paper, an end-to-end deep learning algorithm is adopted to realize the super-resolution of digital holographic imaging. We can conduct high-resolution reconstruction from a low-resolution hologram without any prior information. This method can not only improve the detail information of the image, but also avoid autofocusing and phase recovery in the process of holographic reconstruction.
Improving the Reconstructed Image Quality of the Digital Holography Through Multi-scale Transform

WANG Liang,FENG Shao-tong,NIE Shou-ping,HUANG Fei

DOI: https://doi.org/10.3321/j.issn:1005-0086.2007.05.030

2007-01-01

Abstract:A set of digital image processing algorithms based on multi-scale transform is proposed according to the characteristics of 3-D digital holography.By applying multi-scale principle of wavelet transform and curvelet transform,the zero-order diffraction spot is eliminated,and the speckle noise is reduced effectively while more edge information of object is maintained.As a result,the better quality image is obtained by combining the curvelet transform and digital holography.The experimental result shows that the method is practical and efficient.
Deep Learning-Based Holographic Reconstruction in Digital Holography

Jianglei Di,Kaiqiang Wang,Ying Li,Jianlin Zhao

DOI: https://doi.org/10.1364/dh.2020.htu4b.2

2020-01-01

Abstract:We propose a one-to-two neural network framework for digital holographic reconstruction from a single-wavelength or dual-wavelength digital hologram and also propose a one to one neural network to achieve phase unwrapping.
Multi-Object Deep-Field Digital Holographic Imaging Based on Inverse Cross-Correlation

Jieming Zhao,Zhan Gao,Shengjia Wang,Yuhao Niu,Lin Deng,Ye Sa

DOI: https://doi.org/10.3390/app132011430

2023-01-01

Applied Sciences

Abstract:To address the complexity of small or unique reconstruction distances in digital holography, we propose an inverse cross-correlation-based algorithm for the digital holographic imaging of multiplanar objects with a large depth of field. In this method, a planar output mapping is closely around the objects, and it is established by calculating the image inverse cross-correlation matrix of the reconstructed image at similar reconstruction distances, whereby the object edges serve as the result guide. Combining the search for edge planes with the depth estimation operator, the depth of field of digital holography is improved, thus allowing for a digital holography that is capable of meeting the requirements of the holographic imaging of multiplanar objects. Compared with the traditional depth estimation operator method, the proposed method solves the reconstruction ambiguity problem in multiple planes with a simple optical path, and no additional optical or mechanical devices need to be added, thus greatly improving the reconstruction quality. The numerical calculation results and the experimental results with multiplanar samples validate the effectiveness of the proposed method.
Optimization Of High-Resolution Reconstruction Algorithm For Off-Axis Digital Holographic Imaging

Yibo Wang,Yiwei Liu,Hao Chen,Qiuya Sun,Zhuqing Jiang

DOI: https://doi.org/10.1117/12.2573809

2020-01-01

Abstract:In this paper, an off-axis digital holography multi-frame image super-resolution reconstruction method is presented. Each low-resolution hologram will have small displacement in different directions. The obtained low-resolution holograms are processed iteratively with the super-resolution algorithm to obtain the super-resolution hologram, and then the amplitude image is reconstructed from the super-resolution hologram. The imaging results show that the resolution of the reconstructed image after super-resolution processing is obviously improved, because of increase of high-frequency information in the high-resolution hologram.
Automated phase reconstruction and super-resolution with deep learning in digital holography

Seonghwan Park,Youhyun Kim,Inkyu Moon

DOI: https://doi.org/10.1016/j.optlastec.2024.111030

IF: 4.939

2024-04-17

Optics & Laser Technology

Abstract:Digital holography can provide quantitative phase images that are related to the shape and content of biological samples. In particular, high-resolution phase images contain more important details and information about the sample. However, to obtain a high-resolution phase image, various phase reconstruction processes must be performed, while the finite pixel size of the sensor needs to be overcome. We propose a deep learning model that can obtain high-resolution phase images from low-resolution holograms. The proposed model consists of image translation and super-resolution parts, and performs phase reconstruction and the super-resolution process at the same time. We successfully generated sophisticated phase values that closely resembled real images for three scaling factors of (×2, ×3, and ×4). Comparative evaluations with various deep learning models demonstrated the favorable performance of our proposed model. Multi-scale training was also possible, so it was shown that high-resolution phase images could be generated regardless of the scale factor. The proposed model can automatically generate accurate high-resolution phase images from low-resolution holograms, reducing the cost of digital holography, and providing great benefits to biological sample measurement processes.

optics,physics, applied
End-to-End Reconstruction of Multi-Scale Holograms Based on Cue-Net

Shuo Wang,Xianan Jiang,Xu Liu,Zhao Dong,Ruijing Pei,Huaying Wang

DOI: https://doi.org/10.1016/j.optcom.2022.129079

IF: 2.4

2023-01-01

Optics Communications

Abstract:Digital holography is one of the most widely used label free microscopy techniques in biomedical imaging. It is an important step in hologram reconstruction to recover the missing phase information. A phase recovery method based on deep learning is presented here, which enables the phase and amplitude information of multiple holograms with different out-of-focus distances to be reconstructed quickly, and the multi-scale and large field of view can be reconstructed through the generated network model. We demonstrate the success of this deep learning multi-scale hologram self-focusing reconstruction method by self-focusing the holographic microscopic images of human red blood cells and chicken blood cells. Compared with the existing methods, this method improves the quality of the reconstructed image and the reconstruction speed. Compared with U-NET, the peak signal-to-noise ratio (PSNR) and structural similarity of the reconstructed network output are increased from 18.5 dB and 0.89 to 23.98 dB and 0.97, respectively, which can suppress the influence of speckle noise well.
Deep-learning-aided multi-focal hologram generation

Zhelin Qu,Hao Jiang,Kunpeng Wang,Yunxia Liu,Yang Yang,Chao Feng,Xian Zhao,Jun-Lei Wang

DOI: https://doi.org/10.1016/j.optlastec.2024.112056

IF: 4.939

2024-11-10

Optics & Laser Technology

Abstract:Due to its superior reconfigurability and multifunctionality, computer-generated holograms (CGH) have found extensive utilization in research of light field modulation. Compared with successive light fields such as images, point light fields with multiple foci place higher demands on hologram quality. Moreover, the increasing complexity of the target light field and spatial resolution of holographic displays lead to a corresponding surge in the computation demand for generating high-quality holograms as well as the increasing difficulty to achieve real-time hologram manipulation without compromising the quality. In this study, we demonstrate a deep learning (DL) aided CGH method capable of high fidelity and rapid generation of holograms for multi-foci light field applications by prediction of intrinsic weighting and phase parameter pairs w,θ . Unlike previous DL-based methods that primarily focus on successive light field generation in fixed optical configuration, this approach not only facilitate the initial generation of rapid, high-performance and versatile multi-foci light field in any adjustable optical system, but also allows for subsequent rapid iterative refinements. A lightweight convolutional neural network (CNN) framework is devised to learn the mapping between input light field distributions and w,θ parameters with the linearly corrected root-mean-square error as loss in an unsupervised manner. Our approach stands out in both a non-iterative algorithm with enhanced quality and an iterative algorithm with faster computation speed, adapting to the factual demands while effectively mitigating undesired DL-induced results. The effectiveness of this method is validated through both numerical simulations and experiment demonstrations.

optics,physics, applied
Digital Holographic Reconstruction of Large Objects Using a Convolution Approach and Adjustable Magnification

Jun-chang Li,Patrice Tankam,Zu-jie Peng,Pascal Picart

DOI: https://doi.org/10.1364/ol.34.000572

2009-01-01

Abstract:We present a numerical method for reconstructing large objects using a convolution method with an adjustable magnification. The method is based on the image locations and magnification relations of holography when the illuminating beam is a spherical wavefront. A modified version of the angular spectrum transfer function is proposed that allows the filtering in the spatial frequency spectrum. Experimental results confirm the suitability of the proposed method.
DMEDH: diffusion model-boosted multiplane extrapolation for digital holographic reconstruction

Songyang Gao,Weisheng Xu,Xinyi Wu,Jiawei Liu,Bofei Wang,Tianya Wu,Wenbo Wan,Qiegen Liu

DOI: https://doi.org/10.1364/oe.531147

IF: 3.8

2024-08-22

Optics Express

Abstract:Songyang Gao, Weisheng Xu, Xinyi Wu, Jiawei Liu, Bofei Wang, Tianya Wu, Wenbo Wan, Qiegen Liu Digital holography can reconstruct the amplitude and phase information of the target light field. However, the reconstruction quality is ... [Opt. Express 32, 31920-31938 (2024)]

optics
Application of Deep Learning in Digital Holographic Microscopy

Meng Zhang,Ding Hao,Nie Shouping,Ma Jun,Yuan Caojin

DOI: https://doi.org/10.3788/l0p202158.1811006

2021-01-01

Abstract:Digital holographic microscopy (DHM) has attracted attention in the fields of biological imaging and materials science due to its advantages in high-precision quantitative phase imaging. However, the existence of conjugate images, the problem of phase wrapping, and the limited resolution have always hindered the wide application of DHM. In recent years, deep learning, as a specialized model for data feature extraction in machine learning, has been widely used in the field of optical imaging. In addition to improving imaging efficiency, its potential to solve imaging inverse problems has also been continuously explored by researchers, opening up a new path for the optical imaging. In this paper, we start from the working principle of deep learning applied to DHM, introduces its ideas and important mathematical concepts to solve the inverse problem of optical imaging, and at the same time summarizes the complete implementation process of deep learning. A brief summary of the research progress in recent years of deep learning in holographic reconstruction, auto-focusing and phase recovery, and holographic denoising and super-resolution is given, and summarize the existing problems in this research field and look forward to the development trend of research.
Hough transform-based multi-object autofocusing compressive holography

Yiyi Zhang,Zhengzhong Huang,Shangzhong Jin,Liangcai Cao

DOI: https://doi.org/10.1364/AO.478473

IF: 1.9

2023-01-01

Applied Optics

Abstract:Reconstruction of multiple objects from one hologram can be affected by the focus metric judgment of autofo-cusing. Various segmentation algorithms are applied to obtain a single object in the hologram. Each object is unambiguously reconstructed to acquire its focal position, which produces complicated calculations. Herein, Hough transform (HT)-based multi-object autofocusing compressive holography is presented. The sharpness of each reconstructed image is computed by using a focus metric such as entropy or variance. According to the characteristics of the object, the standard HT is further used for calibration to remove redundant extreme points. The compressive holographic imaging framework with a filter layer can eliminate the inherent noise in in-line reconstruction including cross talk noise of different depth layers, two-order noise, and twin image noise. The proposed method can effectively obtain 3D information on multiple objects and achieve noise elimination by only reconstructing from one hologram. (c) 2023 Optica Publishing Group
Deep DIH : Statistically Inferred Reconstruction of Digital In-Line Holography by Deep Learning

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

DOI: https://doi.org/10.48550/arXiv.2004.12231

2020-04-25

Image and Video Processing

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.
Deep-learning in Digital Holography with Imperfect Data

Zhang Meng,Da Yin,Liqi Ding,Shaotong Feng,Caojin Yuan

DOI: https://doi.org/10.1364/3d.2020.jth4d.1

2020-01-01

Abstract:We propose a digital holographic reconstruction algorithm based on deep learning with imperfect data. Experiments prove that this algorithm can reduce the impact of uncontrollable environmental factors on the dataset and obtain satisfactoiy reconstruction images.
Robust Holographic Reconstruction by Deep Learning with One Frame

Xianfeng Xu,Weilong Luo,Hao Wang,Xinwei Wang

DOI: https://doi.org/10.3390/photonics10101155

IF: 2.536

2023-01-01

Photonics

Abstract:A robust method is proposed to reconstruct images with only one hologram in digital holography by introducing a deep learning (DL) network. The U-net neural network is designed according to DL principles and trained by the image data set collected using phase-shifting digital holography (PSDH). The training data set was established by collecting thousands of reconstructed images using PSDH. The proposed method can complete the holography reconstruction with only a single hologram and then benefits the space bandwidth product and relaxes the storage loads of PSDH. Compared with the results of PSDH, the results of deep learning are immune to most disturbances, including reference tilt, phase-shift errors, and speckle noise. Assisted by a GPU processor, the proposed reconstruction method can reduce the consumption time to almost one percent of the time needed by two-step PSDH. This method is expected to be capable of holography imaging with a single hologram, with high capacity, efficiently in the digital holography applications.
Faithful Digital Holographic Reconstruction Using a Sparse Sensor Array

Zhengzhong Huang,Liangcai Cao

DOI: https://doi.org/10.1063/5.0009191

IF: 4

2020-01-01

Applied Physics Letters

Abstract:The resolution and accuracy of the reconstruction are greatly limited by the field of hologram (FOH) in digital holography. Multi-apertures sampling in the hologram plane can achieve a large FOH. Sparse sensor aperture array uses the synthetic method to achieve higher spatial resolution than any single aperture, which can be used to a large-size scene imaging. However, accurate reconstruction is substantially degraded when there is a large gap between apertures. As a result, the entire scene could not be recovered from FOH. We present a self-restoration method that restores for gaps between sparse apertures. Constraints based on the object energy distribution and absorption characteristics are applied to extract the object function from existing defective reconstruction with a gap. The lost diffraction order can be reproduced and the original object function can be recovered by the iterations between object and hologram. Our iterative process allows a faithful recovery of the gap information to be extracted from the sub-holograms to achieve full-field imaging. The entire complex-value field that contains both the sampling and gap zones in the detection plane is finally retrieved. This faithful reconstruction using a sparse sensor array provides an effectively enlarged hologram for high-resolution digital holography.
High Resolution Digital Holographic Microscopy with A Wide Field of View Based on A Synthetic Aperture Technique and Use of Linear Ccd Scanning

Jianglei Di,Jianlin Zhao,Hongzhen Jiang,Peng Zhang,Qi Fan,Weiwei Sun

DOI: https://doi.org/10.1364/ao.47.005654

IF: 1.9

2008-01-01

Applied Optics

Abstract:Theoretical analysis shows that, to improve the resolution and the range of the field of view of the reconstructed image in digital lensless Fourier transform holography, an effective solution is to increase the area and the pixel number of the recorded digital hologram. A new approach based on the synthetic aperture technique and use of linear CCD scanning is presented to obtain digital holographic images with high resolution and a wide field of view. By using a synthetic aperture technique and linear CCD scanning, we obtained digital lensless Fourier transform holograms with a large area of 3.5 cm x 3.5 cm (5000 x 5000 pixels). The numerical reconstruction of a 4 mm object at a distance of 14 cm by use of a Rayleigh-Sommerfeld integral shows that a theoretically minimum resolvable distance of 2.57 microm can be achieved at a wavelength of 632.8 nm. The experimental results are consistent with the theoretical analysis.
HoloDiffusion: Sparse Digital Holographic Reconstruction via Diffusion Modeling

Liu Zhang,Songyang Gao,Minghao Tong,Yicheng Huang,Zibang Zhang,Wenbo Wan,Qiegen Liu

DOI: https://doi.org/10.3390/photonics11040388

IF: 2.536

2024-04-22

Photonics

Abstract:In digital holography, reconstructed image quality can be primarily limited due to the inability of a single small aperture sensor to cover the entire field of a hologram. The use of multi-sensor arrays in synthetic aperture digital holographic imaging technology contributes to overcoming the limitations of sensor coverage by expanding the area for detection. However, imaging accuracy is affected by the gap size between sensors and the resolution of sensors, especially when dealing with a limited number of sensors. An image reconstruction method is proposed that combines physical constraint characteristics of the imaging object with a score-based diffusion model, aiming to enhance the imaging accuracy of digital holography technology with extremely sparse sensor arrays. Prior information of the sample is learned by the neural network in the diffusion model to obtain a score function, which alternately constrains the iterative reconstruction process with the underlying physical model. The results demonstrate that the structural similarity and peak signal-to-noise ratio of the reconstructed images using this method are higher than the traditional method, along with a strong generalization ability.

optics
Study on Extending the Depth of Field in Reconstructed Image for A Micro Digital Hologram

Yang Jing,Wu Xue-Cheng,Wu Ying-Chun,Yao Long-Chao,Chen Ling-Hong,Qiu Kun-Zan,Cen Ke-Fa

DOI: https://doi.org/10.7498/aps.64.114209

IF: 0.906

2015-01-01

Acta Physica Sinica

Abstract:Digital micro holography offers an in-situ, non-contact and three-dimensional way to explore the microscopic world. However, as it is difficult to focalize the whole object in one single reconstructed image, the application of digital micro holography to cases with a large longitudinal object volume is limited by the microscopes depth of field. By extending the depth of field in reconstructed micro holograms in the wavelet domain, this paper fully takes advantage of numerical reconstruction algorithms to solve this problem. First, a recorded hologram is rebuilt using the wavelet transform approach by setting up an appropriate longitudinal interval to obtain a series of reconstructed hologram planes. Then each plane is decomposed with wavelet into its sub-images of both high and low frequencies. Furthermore, the local variance of the maximum intensity gradients of the high- and low-frequency coefficients is calculated and utilized as the focus criterion. Finally, the image planes are fused into a single one with the depth of field extended to a large extent. The feasibility and robustness of this reconstruction procedure for both continuum and particle fields are investigated. One of the demonstrations is made in an experiment of a tilted continuum:carbon fiber. It is different from most of the previous applications where the interrogated is the particles and where the area involved is parallel to the CCD. The carbon fiber gets successfully reconstructed in three dimensions, and the measurement errors of its diameter are presented together with the reconstruction distances. Another is an experiment of a dispersed particle field:micro transparent particles are generated by an ultrasonic atomizer, for which the reconstruction procedure achieves an extended depth of field. In addition, a numerical model based on generalized Lorenz-Mie theory is used to simulate the holograms of both opaque and transparent particles of 1-15 m in diameter. Variations of the longitudinal location errors with the Fraunhofer number are analyzed, and comparisons are made between the results of opaque and transparent particles. Both the experimental and simulation outcomes show that this reconstruction procedure is a reliable one to acquire an extended-depth-of-field hologram for both the continuum and the dispersed particle fields, and then to accurately measure the objects.

Multi-scale digital holographic reconstruction with deep learning

Multiplane Digital Holography Based on Extrapolation Iterations

Deep Learning Achieves Super-resolution in Digital Holography.

Improving the Reconstructed Image Quality of the Digital Holography Through Multi-scale Transform

Deep Learning-Based Holographic Reconstruction in Digital Holography

Multi-Object Deep-Field Digital Holographic Imaging Based on Inverse Cross-Correlation

Optimization Of High-Resolution Reconstruction Algorithm For Off-Axis Digital Holographic Imaging

Automated phase reconstruction and super-resolution with deep learning in digital holography

End-to-End Reconstruction of Multi-Scale Holograms Based on Cue-Net

Deep-learning-aided multi-focal hologram generation

Digital Holographic Reconstruction of Large Objects Using a Convolution Approach and Adjustable Magnification

DMEDH: diffusion model-boosted multiplane extrapolation for digital holographic reconstruction

Application of Deep Learning in Digital Holographic Microscopy

Hough transform-based multi-object autofocusing compressive holography

Deep DIH : Statistically Inferred Reconstruction of Digital In-Line Holography by Deep Learning

Deep-learning in Digital Holography with Imperfect Data

Robust Holographic Reconstruction by Deep Learning with One Frame

Faithful Digital Holographic Reconstruction Using a Sparse Sensor Array

High Resolution Digital Holographic Microscopy with A Wide Field of View Based on A Synthetic Aperture Technique and Use of Linear Ccd Scanning

HoloDiffusion: Sparse Digital Holographic Reconstruction via Diffusion Modeling

Study on Extending the Depth of Field in Reconstructed Image for A Micro Digital Hologram