Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2006.09410

Fast Correlated-Photon Imaging Enhanced by Deep Learning

Zhanming Li, Shibao Wu, Jun Gao, Heng Zhou, Zeng‐Quan Yan, Ruo-Jing Ren, Siyuan Yin, Xian‐Min Jin

Published Jun 16, 2020• 1 citation
arXivHTMLPDF

Abstract

Correlated photon pairs, carrying strong quantum correlations, have been harnessed to bring quantum advantages to various fields from biological imaging to range finding. Such inherent non-classical properties support extracting more valid signals to build photon-limited images even in low flux-level, where the shot noise becomes dominant as light source decreases to single-photon level. Optimization by numerical reconstruction algorithms is possible but require thousands of photon-sparse frames, thus unavailable in real time. Here, we present an experimental fast correlated-photon imaging enhanced by deep learning, showing an intelligent computational strategy to discover deeper structure in big data. Convolutional neural network is found being able to efficiently solve image inverse problems associated with strong shot noise and background noise (electronic noise, scattered light). Our results fill the key gap in incompatibility between imaging speed and image quality by pushing low-light imaging technique to the regime of real-time and single-photon level, opening up an avenue to deep leaning-enhanced quantum imaging for real-life applications.

DOI
10.48550/arxiv.2006.09410

Related papers

Fast correlated-photon imaging enhanced by deep learning

Zhanming Li, Shibao Wu, Jun Gao +5 more

Quantum imaging using photon pairs with strong quantum correlations has been harnessed to bring quantum advantages to various fields from biological imaging to range finding. Such inherent non-classical properties support the extraction of more valid signals to build photon-limited images, even in low-light conditions where the shot noise becomes dominant as light decreases to a single-photon level. Numerical optimization algorithms are possible but require thousands of photon-sparse frames, and they are thus unavailable in real time. We demonstrate fast correlated-photon imaging enhanced by deep learning as an intelligent computational strategy to discover a deeper structure in big data. Our work verifies that a convolutional neural network can efficiently solve inverse imaging problems associated with strong shot noise and background noise (electronic noise, scattered light). Our results show that we can overcome limitations due to the trade-off between imaging speed and image quality by pushing the low-light imaging technique to the single-photon level in real time, which enables deep-learning-enhanced quantum imaging for real-life applications.

Optica•Jan 13, 2021•35 citations

Smart Quantum Imaging

Ebrahim Karimi

The use of quantum properties of light has paved new avenues in the field of imaging. Quantum illumination and ghost-imaging protocols have shown exciting promise in addressing the issues inherent in classical imaging, such as background noise and a high level of required illumination. Moreover, quantum metrological schemes analyzing the modal content of light would address the resolution limits of direct imaging, promising enhancement beyond the Rayleigh limit. Augmenting these novel techniques with machine learning, particularly deep learning architectures such as Convolutional Neural Networks (CNNs), one strives to see significant improvements in image reconstruction and object identification in quantum imaging protocols. In my talk, I present recent progress and development on reconstructing the image of different objects with quantum sources, employing a high degree of spatial correlation between photon pairs and autoencoder deep learning architecture.

Photonics for Quantum (PfQ)•Jul 15, 2021•0 citations

Classification of quantum correlation using deep learning

Shibao Wu, Zhanming Li, Jun Gao +3 more

Quantum correlation, as an intrinsic property of quantum mechanics, has been widely employed to test the fundamental physical principles and explore the quantum-enhanced technologies. However, such correlation would be drowned and even destroyed in the conditions of high levels of loss and noise, which drops into the classical realm and renders quantum advantage ineffective. Especially in low light conditions, conventional linear classifiers are unable to extract and distinguish quantum and classical correlations with high accuracy. Here we experimentally demonstrate the classification of quantum correlation using deep learning to meet the challenge in the quantum imaging scheme. We design the convolutional neural network to learn and classify the correlated photons efficiently with only 0.1 signal photons per pixel. We show that decreasing signal intensity further weakens the correlation and makes an accurate linear classification impossible, while the deep learning method has a strong robustness of such task with the accuracy of 99.99%. These results open up a new perspective to optimize the quantum correlation in low light conditions, representing a step towards diverse applications in quantum-enhanced measurement scenarios, such as super-resolution microscope, quantum illumination, etc.

Optics Express•Jan 9, 2023•7 citations

Time-efficient non-degenerate ghost imaging powered by deep learning

Chané Moodley, Bereneice Sephton, Valeria Rodríguez-Fajardo +1 more

Quantum ghost imaging utilizes entangled photon pairs to enable an alternative image acquisition method. Information from either one of the photons does not allow for image reconstruction, however the image can be reconstructed by utilising the correlations that exist between the photon pair. Interestingly, these photon pairs can be either degenerate or non-degenerate in nature. Non-degenerate ghost imaging offers the ability to image with wavelength bandwidths where spatially resolving detectors are impractical, ineffective or expensive. Due to the scanning nature of spatially resolving detectors and the inherent low light levels of quantum experiments, imaging speeds are rather unsatisfactory. To overcome this limitation, we propose a two-step deep learning approach to establish an optimal early stopping point, tested on a non-degenerate system. In step one, we enhance the reconstructed image after each measurement by a deep convolution auto-encoder, followed by step two where a classifier is used to recognise the image. We achieved a recognition confidence of 75% at 20% of the image reconstruction time, hence reducing the image reconstruction time 5-fold while maintaining the image information. This, therefore, leads to a faster, more efficient image acquisition technique. Although tested on a non-degenerate system, our procedure can be extended to many such systems that are of quantum nature. We believe that this two-step deep learning approach will prove valuable to the community who are focusing their efforts on time-efficient ghost imaging.

Quantum Communications and Quantum Imaging•Jul 30, 2021•1 citation

Surpassing the standard quantum limit of optical imaging via deep learning

Miao Cai, Zhixiang Li, Haodong Wu +8 more

The sensitivity of optical measurement is ultimately constrained by the shot noise to the standard quantum limit.It has become a common concept that beating this limit requires quantum resources.A deep-learning neural network free of quantum principle has the capability of removing classical noise from images, but it is unclear in reducing quantum noise.In a coincidence-imaging experiment, we show that quantum-resource-free deep learning can be exploited to surpass the standard quantum limit via the photon-number-dependent nonlinear feedback during training.Using an effective classical light with photon flux of about 9 × 10 4 photons per second, our deep-learning-based scheme achieves a 14 dB improvement in signal-to-noise ratio with respect to the standard quantum limit.

Chinese Optics Letters•Jan 1, 2023•2 citations

Imaging at the quantum limit with convolutional neural networks

Andrew H. Proppe, Aaron Z. Goldberg, Guillaume Thekkadath +8 more

Deep neural networks have been shown to achieve exceptional performance for computer vision tasks like image recognition, segmentation, and reconstruction or denoising. Here, we evaluate the ultimate performance limits of deep convolutional neural network models for image reconstruction, by comparing them against the standard quantum limit set by shot-noise and the Heisenberg limit on precision. We train U-Net models on images of natural objects illuminated with coherent states of light, and find that the average mean-squared error of the reconstructions can surpass the standard quantum limit, and in some cases reaches the Heisenberg limit. Further, we train models on well-parameterized images for which we can calculate the quantum Cramér-Rao bound to determine the minimum possible measurable variance of an estimated parameter for a given probe state. We find the mean-squared error of the model predictions reaches these bounds calculated for the parameters, across a variety of parameterized images. These results suggest that deep convolutional neural networks can learn to become the optimal estimators allowed by the laws of physics, performing parameter estimation and image reconstruction at the ultimate possible limits of precision for the case of classical illumination of the object.

arXiv:2506.13488•Jun 16, 2025•0 citations
Correlated photon pairs, carrying strong quantum correlations, have been harnessed to bring quantum advantages to various fields from biological imaging to range finding. Such inherent non-classical properties support extracting more valid signals to build photon-limited images even in low flux-level, where the shot noise becomes dominant as light source decreases to single-photon level. Optimization by numerical reconstruction algorithms is possible but require thousands of photon-sparse frames, thus unavailable in real time. Here, we present an experimental fast correlated-photon imaging enhanced by deep learning, showing an intelligent computational strategy to discover deeper structure in big data. Convolutional neural network is found being able to efficiently solve image inverse problems associated with strong shot noise and background noise (electronic noise, scattered light). Our results fill the key gap in incompatibility between imaging speed and image quality by pushing low-light imaging technique to the regime of real-time and single-photon level, opening up an avenue to deep leaning-enhanced quantum imaging for real-life applications.
Quantum imaging using photon pairs with strong quantum correlations has been harnessed to bring quantum advantages to various fields from biological imaging to range finding. Such inherent non-classical properties support the extraction of more valid signals to build photon-limited images, even in low-light conditions where the shot noise becomes dominant as light decreases to a single-photon level. Numerical optimization algorithms are possible but require thousands of photon-sparse frames, and they are thus unavailable in real time. We demonstrate fast correlated-photon imaging enhanced by deep learning as an intelligent computational strategy to discover a deeper structure in big data. Our work verifies that a convolutional neural network can efficiently solve inverse imaging problems associated with strong shot noise and background noise (electronic noise, scattered light). Our results show that we can overcome limitations due to the trade-off between imaging speed and image quality by pushing the low-light imaging technique to the single-photon level in real time, which enables deep-learning-enhanced quantum imaging for real-life applications.
The use of quantum properties of light has paved new avenues in the field of imaging. Quantum illumination and ghost-imaging protocols have shown exciting promise in addressing the issues inherent in classical imaging, such as background noise and a high level of required illumination. Moreover, quantum metrological schemes analyzing the modal content of light would address the resolution limits of direct imaging, promising enhancement beyond the Rayleigh limit. Augmenting these novel techniques with machine learning, particularly deep learning architectures such as Convolutional Neural Networks (CNNs), one strives to see significant improvements in image reconstruction and object identification in quantum imaging protocols. In my talk, I present recent progress and development on reconstructing the image of different objects with quantum sources, employing a high degree of spatial correlation between photon pairs and autoencoder deep learning architecture.
Quantum correlation, as an intrinsic property of quantum mechanics, has been widely employed to test the fundamental physical principles and explore the quantum-enhanced technologies. However, such correlation would be drowned and even destroyed in the conditions of high levels of loss and noise, which drops into the classical realm and renders quantum advantage ineffective. Especially in low light conditions, conventional linear classifiers are unable to extract and distinguish quantum and classical correlations with high accuracy. Here we experimentally demonstrate the classification of quantum correlation using deep learning to meet the challenge in the quantum imaging scheme. We design the convolutional neural network to learn and classify the correlated photons efficiently with only 0.1 signal photons per pixel. We show that decreasing signal intensity further weakens the correlation and makes an accurate linear classification impossible, while the deep learning method has a strong robustness of such task with the accuracy of 99.99%. These results open up a new perspective to optimize the quantum correlation in low light conditions, representing a step towards diverse applications in quantum-enhanced measurement scenarios, such as super-resolution microscope, quantum illumination, etc.
Quantum ghost imaging utilizes entangled photon pairs to enable an alternative image acquisition method. Information from either one of the photons does not allow for image reconstruction, however the image can be reconstructed by utilising the correlations that exist between the photon pair. Interestingly, these photon pairs can be either degenerate or non-degenerate in nature. Non-degenerate ghost imaging offers the ability to image with wavelength bandwidths where spatially resolving detectors are impractical, ineffective or expensive. Due to the scanning nature of spatially resolving detectors and the inherent low light levels of quantum experiments, imaging speeds are rather unsatisfactory. To overcome this limitation, we propose a two-step deep learning approach to establish an optimal early stopping point, tested on a non-degenerate system. In step one, we enhance the reconstructed image after each measurement by a deep convolution auto-encoder, followed by step two where a classifier is used to recognise the image. We achieved a recognition confidence of 75% at 20% of the image reconstruction time, hence reducing the image reconstruction time 5-fold while maintaining the image information. This, therefore, leads to a faster, more efficient image acquisition technique. Although tested on a non-degenerate system, our procedure can be extended to many such systems that are of quantum nature. We believe that this two-step deep learning approach will prove valuable to the community who are focusing their efforts on time-efficient ghost imaging.
The sensitivity of optical measurement is ultimately constrained by the shot noise to the standard quantum limit.It has become a common concept that beating this limit requires quantum resources.A deep-learning neural network free of quantum principle has the capability of removing classical noise from images, but it is unclear in reducing quantum noise.In a coincidence-imaging experiment, we show that quantum-resource-free deep learning can be exploited to surpass the standard quantum limit via the photon-number-dependent nonlinear feedback during training.Using an effective classical light with photon flux of about 9 × 10 4 photons per second, our deep-learning-based scheme achieves a 14 dB improvement in signal-to-noise ratio with respect to the standard quantum limit.
Deep neural networks have been shown to achieve exceptional performance for computer vision tasks like image recognition, segmentation, and reconstruction or denoising. Here, we evaluate the ultimate performance limits of deep convolutional neural network models for image reconstruction, by comparing them against the standard quantum limit set by shot-noise and the Heisenberg limit on precision. We train U-Net models on images of natural objects illuminated with coherent states of light, and find that the average mean-squared error of the reconstructions can surpass the standard quantum limit, and in some cases reaches the Heisenberg limit. Further, we train models on well-parameterized images for which we can calculate the quantum Cramér-Rao bound to determine the minimum possible measurable variance of an estimated parameter for a given probe state. We find the mean-squared error of the model predictions reaches these bounds calculated for the parameters, across a variety of parameterized images. These results suggest that deep convolutional neural networks can learn to become the optimal estimators allowed by the laws of physics, performing parameter estimation and image reconstruction at the ultimate possible limits of precision for the case of classical illumination of the object.