Efficient high-resolution microscopic ghost imaging via sequenced speckle illumination and deep learning from a single noisy image

This study presents a novel approach for achieving high-quality and large-scale microscopic ghost imaging by integrating deep learning-based denoising with computational ghost imaging techniques. By utilizing sequenced random speckle patterns of optimized sizes, we reconstructed large noisy images w...

Full description

Saved in:
Bibliographic Details
Main Authors: Sukyoon Oh, Tong Tian, Zhe Sun, Christian Spielmann
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-06-01
Series:Advanced Optical Technologies
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/aot.2025.1583836/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849683634068389888
author Sukyoon Oh
Sukyoon Oh
Tong Tian
Tong Tian
Zhe Sun
Christian Spielmann
Christian Spielmann
author_facet Sukyoon Oh
Sukyoon Oh
Tong Tian
Tong Tian
Zhe Sun
Christian Spielmann
Christian Spielmann
author_sort Sukyoon Oh
collection DOAJ
description This study presents a novel approach for achieving high-quality and large-scale microscopic ghost imaging by integrating deep learning-based denoising with computational ghost imaging techniques. By utilizing sequenced random speckle patterns of optimized sizes, we reconstructed large noisy images with fewer patterns while successfully resolving fine details as small as 2.2 μm on a USAF resolution target. To enhance image quality, we incorporated the Deep Neural Network-based Noise2Void (N2V) model, which effectively denoises ghost images without requiring a reference image or a large dataset. By applying the N2V model to a single noisy ghost image, we achieved significant noise reduction, leading to high-resolution and high-quality reconstructions with low computational resources. This method resulted in an average Structural Similarity Index (SSIM) improvement of over 324% and a resolution enhancement exceeding 33% across various target images. The proposed approach proves highly effective in enhancing the clarity and structural integrity of even very low-quality ghost images, paving the way for more efficient and practical implementations of ghost imaging in microscopic applications.
format Article
id doaj-art-aec437dfae4e4e298f47e2c12378c5cb
institution DOAJ
issn 2192-8584
language English
publishDate 2025-06-01
publisher Frontiers Media S.A.
record_format Article
series Advanced Optical Technologies
spelling doaj-art-aec437dfae4e4e298f47e2c12378c5cb2025-08-20T03:23:46ZengFrontiers Media S.A.Advanced Optical Technologies2192-85842025-06-011410.3389/aot.2025.15838361583836Efficient high-resolution microscopic ghost imaging via sequenced speckle illumination and deep learning from a single noisy imageSukyoon Oh0Sukyoon Oh1Tong Tian2Tong Tian3Zhe Sun4Christian Spielmann5Christian Spielmann6Institute of Optics and Quantum Electronics, Abbe Center of Photonics, Friedrich Schiller University, Jena, GermanyGSI Helmholtz Centre for Heavy Ion Research, Darmstadt, GermanyInstitute of Optics and Quantum Electronics, Abbe Center of Photonics, Friedrich Schiller University, Jena, GermanyGSI Helmholtz Centre for Heavy Ion Research, Darmstadt, GermanySchool of Artificial Intelligence, OPtics and ElectroNics (iOPEN), Northwestern Polytechnical University, Xi’an, Shaanxi, ChinaInstitute of Optics and Quantum Electronics, Abbe Center of Photonics, Friedrich Schiller University, Jena, GermanyGSI Helmholtz Centre for Heavy Ion Research, Darmstadt, GermanyThis study presents a novel approach for achieving high-quality and large-scale microscopic ghost imaging by integrating deep learning-based denoising with computational ghost imaging techniques. By utilizing sequenced random speckle patterns of optimized sizes, we reconstructed large noisy images with fewer patterns while successfully resolving fine details as small as 2.2 μm on a USAF resolution target. To enhance image quality, we incorporated the Deep Neural Network-based Noise2Void (N2V) model, which effectively denoises ghost images without requiring a reference image or a large dataset. By applying the N2V model to a single noisy ghost image, we achieved significant noise reduction, leading to high-resolution and high-quality reconstructions with low computational resources. This method resulted in an average Structural Similarity Index (SSIM) improvement of over 324% and a resolution enhancement exceeding 33% across various target images. The proposed approach proves highly effective in enhancing the clarity and structural integrity of even very low-quality ghost images, paving the way for more efficient and practical implementations of ghost imaging in microscopic applications.https://www.frontiersin.org/articles/10.3389/aot.2025.1583836/fullghost imaging (GI)deep learningNoise2Voidsingle pixel imagingmicroscopydenoising
spellingShingle Sukyoon Oh
Sukyoon Oh
Tong Tian
Tong Tian
Zhe Sun
Christian Spielmann
Christian Spielmann
Efficient high-resolution microscopic ghost imaging via sequenced speckle illumination and deep learning from a single noisy image
Advanced Optical Technologies
ghost imaging (GI)
deep learning
Noise2Void
single pixel imaging
microscopy
denoising
title Efficient high-resolution microscopic ghost imaging via sequenced speckle illumination and deep learning from a single noisy image
title_full Efficient high-resolution microscopic ghost imaging via sequenced speckle illumination and deep learning from a single noisy image
title_fullStr Efficient high-resolution microscopic ghost imaging via sequenced speckle illumination and deep learning from a single noisy image
title_full_unstemmed Efficient high-resolution microscopic ghost imaging via sequenced speckle illumination and deep learning from a single noisy image
title_short Efficient high-resolution microscopic ghost imaging via sequenced speckle illumination and deep learning from a single noisy image
title_sort efficient high resolution microscopic ghost imaging via sequenced speckle illumination and deep learning from a single noisy image
topic ghost imaging (GI)
deep learning
Noise2Void
single pixel imaging
microscopy
denoising
url https://www.frontiersin.org/articles/10.3389/aot.2025.1583836/full
work_keys_str_mv AT sukyoonoh efficienthighresolutionmicroscopicghostimagingviasequencedspeckleilluminationanddeeplearningfromasinglenoisyimage
AT sukyoonoh efficienthighresolutionmicroscopicghostimagingviasequencedspeckleilluminationanddeeplearningfromasinglenoisyimage
AT tongtian efficienthighresolutionmicroscopicghostimagingviasequencedspeckleilluminationanddeeplearningfromasinglenoisyimage
AT tongtian efficienthighresolutionmicroscopicghostimagingviasequencedspeckleilluminationanddeeplearningfromasinglenoisyimage
AT zhesun efficienthighresolutionmicroscopicghostimagingviasequencedspeckleilluminationanddeeplearningfromasinglenoisyimage
AT christianspielmann efficienthighresolutionmicroscopicghostimagingviasequencedspeckleilluminationanddeeplearningfromasinglenoisyimage
AT christianspielmann efficienthighresolutionmicroscopicghostimagingviasequencedspeckleilluminationanddeeplearningfromasinglenoisyimage