Color Fourier single-pixel imaging(FSI)enables efficient spectral and spatial imaging.Here,we propose a Fourier single-pixel imaging scheme with a random color filter array(FSI-RCFA).The proposed method employs a rand...Color Fourier single-pixel imaging(FSI)enables efficient spectral and spatial imaging.Here,we propose a Fourier single-pixel imaging scheme with a random color filter array(FSI-RCFA).The proposed method employs a random color filter array(RCFA)to modulate Fourier patterns.A three-step phase-shifting technique reconstructs the Fourier spectrum,followed by an RCFA-based demosaicing algorithm to recover color images.Compared to traditional color FSI based on Bayer color filter array schemes(FSI-BCFA),our approach achieves superior separation between chrominance and luminance components in the frequency domain.Simulation results demonstrate that the FSI-RCFA method achieves a lower mean squared error(MSE),a higher peak signal-to-noise ratio(PSNR),and superior noise resistance compared to FSI-BCFA,while enabling direct single-channel pixel measurements for targeted applications such as agricultural defect detection.展开更多
Single-pixel imaging(SPI)enables efficient sensing in challenging conditions.However,the requirement for numerous samplings constrains its practicality.We address the challenge of high-quality SPI reconstruction at ul...Single-pixel imaging(SPI)enables efficient sensing in challenging conditions.However,the requirement for numerous samplings constrains its practicality.We address the challenge of high-quality SPI reconstruction at ultra-low sampling rates.We develop an alternative optimization with physics and a data-driven diffusion network(APD-Net).It features alternative optimization driven by the learned task-agnostic natural image prior and the task-specific physics prior.During the training stage,APD-Net harnesses the power of diffusion models to capture data-driven statistics of natural signals.In the inference stage,the physics prior is introduced as corrective guidance to ensure consistency between the physics imaging model and the natural image probability distribution.Through alternative optimization,APD-Net reconstructs data-efficient,high-fidelity images that are statistically and physically compliant.To accelerate reconstruction,initializing images with the inverse SPI physical model reduces the need for reconstruction inference from 100 to 30 steps.Through both numerical simulations and real prototype experiments,APD-Net achieves high-quality,full-color reconstructions of complex natural images at a low sampling rate of 1%.In addition,APD-Net’s tuning-free nature ensures robustness across various imaging setups and sampling rates.Our research offers a broadly applicable approach for various applications,including but not limited to medical imaging and industrial inspection.展开更多
Existing single-pixel imaging(SPI)and sensing techniques suffer from poor reconstruction quality and heavy computation cost,limiting their widespread application.To tackle these challenges,we propose a large-scale sin...Existing single-pixel imaging(SPI)and sensing techniques suffer from poor reconstruction quality and heavy computation cost,limiting their widespread application.To tackle these challenges,we propose a large-scale single-pixel imaging and sensing(SPIS)technique that enables high-quality megapixel SPI and highly efficient image-free sensing with a low sampling rate.Specifically,we first scan and sample the entire scene using small-size optimized patterns to obtain information-coupled measurements.Compared with the conventional full-sized patterns,small-sized optimized patterns achieve higher imaging fidelity and sensing accuracy with 1 order of magnitude fewer pattern parameters.Next,the coupled measurements are processed through a transformer-based encoder to extract high-dimensional features,followed by a task-specific plugand-play decoder for imaging or image-free sensing.Considering that the regions with rich textures and edges are more difficult to reconstruct,we use an uncertainty-driven self-adaptive loss function to reinforce the network’s attention to these regions,thereby improving the imaging and sensing performance.Extensive experiments demonstrate that the reported technique achieves 24.13 dB megapixel SPI at a sampling rate of 3%within 1 s.In terms of sensing,it outperforms existing methods by 12%on image-free segmentation accuracy and achieves state-of-the-art image-free object detection accuracy with an order of magnitude less data bandwidth.展开更多
Single-pixel imaging(SPI)is a prominent scattering media imaging technique that allows image transmission via one-dimensional detection under structured illumination,with applications spanning from long-range imaging ...Single-pixel imaging(SPI)is a prominent scattering media imaging technique that allows image transmission via one-dimensional detection under structured illumination,with applications spanning from long-range imaging to microscopy.Recent advancements leveraging deep learning(DL)have significantly improved SPI performance,especially at low compression ratios.However,most DL-based SPI methods proposed so far rely heavily on extensive labeled datasets for supervised training,which are often impractical in real-world scenarios.Here,we propose an unsupervised learningenabled label-free SPI method for resilient information transmission through unknown dynamic scattering media.Additionally,we introduce a physics-informed autoencoder framework to optimize encoding schemes,further enhancing image quality at low compression ratios.Simulation and experimental results demonstrate that high-efficiency data transmission with structural similarity exceeding 0.9 is achieved through challenging turbulent channels.Moreover,experiments demonstrate that in a 5 m underwater dynamic turbulent channel,USAF target imaging quality surpasses traditional methods by over 13 dB.The compressive encoded transmission of 720×720 resolution video exceeding 30 seconds with great fidelity is also successfully demonstrated.These preliminary results suggest that our proposed method opens up a new paradigm for resilient information transmission through unknown dynamic scattering media and holds potential for broader applications within many other scattering media imaging technologies.展开更多
In recent years,deep learning has been introduced into the field of Single-pixel imaging(SPI),garnering significant attention.However,conventional networks still exhibit limitations in preserving image details.To addr...In recent years,deep learning has been introduced into the field of Single-pixel imaging(SPI),garnering significant attention.However,conventional networks still exhibit limitations in preserving image details.To address this issue,we integrate Large Kernel Convolution(LKconv)into the U-Net framework,proposing an enhanced network structure named U-LKconv network,which significantly enhances the capability to recover image details even under low sampling conditions.展开更多
We propose a method of complex-amplitude Fourier single-pixel imaging(CFSI)with coherent structured illumination to acquire both the amplitude and phase of an object.In the proposed method,an object is illustrated by ...We propose a method of complex-amplitude Fourier single-pixel imaging(CFSI)with coherent structured illumination to acquire both the amplitude and phase of an object.In the proposed method,an object is illustrated by a series of coherent structured light fields,which are generated by a phase-only spatial light modulator,the complex Fourier spectrum of the object can be acquired sequentially by a single-pixel photodetector.Then the desired complex-amplitude image can be retrieved directly by applying an inverse Fourier transform.We experimentally implemented this CFSI with several different types of objects.The experimental results show that the proposed method provides a promising complex-amplitude imaging approach with high quality and a stable configuration.Thus,it might find broad applications in optical metrology and biomedical science.展开更多
Single-pixel imaging(SPI)can transform 2D or 3D image data into 1D light signals,which offers promising prospects for image compression and transmission.However,during data communication these light signals in public ...Single-pixel imaging(SPI)can transform 2D or 3D image data into 1D light signals,which offers promising prospects for image compression and transmission.However,during data communication these light signals in public channels will easily draw the attention of eavesdroppers.Here,we introduce an efficient encryption method for SPI data transmission that uses the 3D Arnold transformation to directly disrupt 1D single-pixel light signals and utilizes the elliptic curve encryption algorithm for key transmission.This encryption scheme immediately employs Hadamard patterns to illuminate the scene and then utilizes the 3D Arnold transformation to permutate the 1D light signal of single-pixel detection.Then the transformation parameters serve as the secret key,while the security of key exchange is guaranteed by an elliptic curve-based key exchange mechanism.Compared with existing encryption schemes,both computer simulations and optical experiments have been conducted to demonstrate that the proposed technique not only enhances the security of encryption but also eliminates the need for complicated pattern scrambling rules.Additionally,this approach solves the problem of secure key transmission,thus ensuring the security of information and the quality of the decrypted images.展开更多
We propose a single-pixel imaging(SPI)method to achieve a higher-resolution image via the Hadamard transform matrix.Unlike traditional SPI schemes,this new method recovers images by correlating single-pixel signals wi...We propose a single-pixel imaging(SPI)method to achieve a higher-resolution image via the Hadamard transform matrix.Unlike traditional SPI schemes,this new method recovers images by correlating single-pixel signals with synchronized transformed patterns of Hadamard bases that are actually projected onto the digital micromirror device.Each transform pattern is obtained through the inverse Fourier transform of the pattern acquired by Gaussian filtering of each Hadamard basis in the frequency domain.The proposed scheme is based on a typical SPI experimental setup and does not add any hardware complexity,enabling the transformation of Hadamard matrices and image reconstruction through data processing alone.Therefore,this approach could be considered as an alternative option for achieving fast SPI in a diffraction-limited imaging system,without the need for additional hardware.展开更多
The single-pixel imaging(SPI) technique is able to capture two-dimensional(2 D) images without conventional array sensors by using a photodiode. As a novel scheme, Fourier single-pixel imaging(FSI) has been proven cap...The single-pixel imaging(SPI) technique is able to capture two-dimensional(2 D) images without conventional array sensors by using a photodiode. As a novel scheme, Fourier single-pixel imaging(FSI) has been proven capable of reconstructing high-quality images. Due to the fact that the Fourier basis patterns(also known as grayscale sinusoidal patterns)cannot be well displayed on the digital micromirror device(DMD), a fast FSI system is proposed to solve this problem by binarizing Fourier pattern through a dithering algorithm. However, the traditional dithering algorithm leads to low quality as the extra noise is inevitably induced in the reconstructed images. In this paper, we report a better dithering algorithm to binarize Fourier pattern, which utilizes the Sierra–Lite kernel function by a serpentine scanning method. Numerical simulation and experiment demonstrate that the proposed algorithm is able to achieve higher quality under different sampling ratios.展开更多
A two-stage training method is proposed to enhance imaging quality and reduce reconstruction time in datadriven single-pixel imaging(SPI)under undersampling conditions.This approach leverages a deep learning algorithm...A two-stage training method is proposed to enhance imaging quality and reduce reconstruction time in datadriven single-pixel imaging(SPI)under undersampling conditions.This approach leverages a deep learning algorithm to simulate single-pixel detection and image reconstruction.During the initial training stage,an L2 regularization constraint is imposed on convolution modulation patterns to determine the optimal initial network weights.In the subsequent stage,a coupled deep learning method integrating coded-aperture design and SPI is adopted,which utilizes backpropagation of the loss function to iteratively optimize both the binarized modulation patterns and imaging network parameters.By reducing the binarization errors introduced by the dithering algorithm,this approach improves the quality of data-driven SPI.Compared with traditional deep-learning SPI methods,the proposed method significantly reduces computational complexity,resulting in accelerated image reconstruction.Experimental and simulation results demonstrate the advantages of the method,including high imaging quality,short image reconstruction time,and simplified training.For an image size of 64×64 pixels and 10%sampling rate,the proposed method achieves a peak signal-to-noise ratio of 23.22 dB,structural similarity index of 0.76,and image reconstruction time of approximately 2.57×10^(−4) seconds.展开更多
Photon-level single-pixel imaging overcomes the reliance of traditional imaging techniques on large-scale array detectors,offering the advantages such as high sensitivity,high resolution,and efficient photon utilizati...Photon-level single-pixel imaging overcomes the reliance of traditional imaging techniques on large-scale array detectors,offering the advantages such as high sensitivity,high resolution,and efficient photon utilization.In this paper,we propose a photon-level dynamic feature single-pixel imaging method,leveraging the frequency domain sparsity of the object's dynamic features to construct a compressed measurement system through discrete random photon detection.In the experiments,we successfully captured 167 and 200 Hz featured frequencies and achieved high-quality image reconstruction with a data compression ratio of 20%.Our approach introduces a new detection dimension,significantly expanding the applications of photon-level single-pixel imaging in practical scenarios.展开更多
Existing two-step Fourier single-pixel imaging (FSPI) suffers from low noise-robustness, and three-step FSPI and four-step FSPI improve the noise-robustness but at the cost of more measurements. In this Letter, we pro...Existing two-step Fourier single-pixel imaging (FSPI) suffers from low noise-robustness, and three-step FSPI and four-step FSPI improve the noise-robustness but at the cost of more measurements. In this Letter, we propose a method to improve the noise-robustness of two-step FSPI without additional illumination patterns or measurements. In the proposed method, the measurements from base patterns are replaced by the average values of the measurement from two sets of phase-shift patterns. Thus, the imaging degradation caused by the noise in the measurements from base patterns can be avoided, and more reliable Fourier spectral coefficients are obtained. The imaging quality of the proposed robust two-step FSPI is similar to those of three-step FSPI and four-step FSPI. Simulations and experimental results validate the effectiveness of the proposed method.展开更多
Neutron-induced gamma-ray imaging is a spectroscopic technique that uses characteristic gamma rays to infer the elemental distribution of an object.Currently,this technique requires the use of large facilities to supp...Neutron-induced gamma-ray imaging is a spectroscopic technique that uses characteristic gamma rays to infer the elemental distribution of an object.Currently,this technique requires the use of large facilities to supply a high neutron flux and a time-consuming detection procedure involving direct collimating measurements.In this study,a new method based on low neutron flux was proposed.A single-pixel gamma-ray detector combined with random pattern gamma-ray masks was used to measure the characteristic gamma rays emitted from the sample.Images of the elemental distribution in the sample,comprising 30×30 pixels,were reconstructed using the maximum-likelihood expectation-maximization algorithm.The results demonstrate that the elemental imaging of the sample can be accurately determined using this method.The proposed approach,which eliminates the need for high neutron flux and scanning measurements,can be used for in-field imaging applications.展开更多
Background:Sarcomatoid carcinoma of the ureter(SCU)is a highly aggressive and relatively uncommon malignant tumor of the urinary tract.Its frequency is quite low,and its prognosis is very bad when compared to other ca...Background:Sarcomatoid carcinoma of the ureter(SCU)is a highly aggressive and relatively uncommon malignant tumor of the urinary tract.Its frequency is quite low,and its prognosis is very bad when compared to other cancers of the urinary system.SCU clinical reports are still hard to come by.MRI and PEI/CT imaging of ureteral sarcomatoid cancer is presented in this case to promote diagnostic awareness and comprehension of the imaging characteristics of this uncommon illness.Method:The patient had ureteral sarcomatoid cancer,which was verified by pathological investigation after ureteroscopic biopsy.The patient’s clinical information,imaging results,surgical outcomes,and pathological findings were gathered.A retrospective study was carried out in combinationwith pertinent national and international literature.Results:An 84-year-old female patient was admitted for“left flank discomfort lasting over one month.”MRI revealed an irregular soft tissue mass in the middle-lower segment of the left ureter.T2-weighted imaging showed an unevenly slightly hyperintense signal.Diffusion-weighted imaging demonstrated restricted diffusion.Contrastenhanced imaging exhibited heterogeneous enhancement.PET/CT demonstrated significantly increased fluorodeoxyglucose metabolism in the mass with secondary left upper urinary tract obstruction.Concurrent findings included a solitary metastatic lesion in hepatic segment S6 and multiple lymph node metastases along the left common iliac and external iliac arteries.Preoperative diagnosis suggested a malignant tumor of the ureter.The patient underwent left nephroureteroscopy with biopsy,and the postoperative pathological diagnosis was ureteral sarcomatoid carcinoma.Conclusion:Ureteral sarcomatoid carcinoma is a rare,highly malignant,and aggressive tumor with nonspecific imaging features,typically presenting as an invasively growing mass.Diagnosis relies on postoperative pathology and immunohistochemical examination.MRI and PET/CT scans are valuable for preoperative localization and characterization,tumor staging,treatment planning,and postoperative follow-up.The prognosis is extremely negative.The main treatment option is radical surgery,although constant monitoring is necessary since early recurrence and metastases are frequent after surgery.展开更多
The level of glutathione(GSH)is significantly associated with numerous pathological processes,thus,real-time detection of the GSH level is of significance for early diagnosis of GSH-related diseases.Herein,we develope...The level of glutathione(GSH)is significantly associated with numerous pathological processes,thus,real-time detection of the GSH level is of significance for early diagnosis of GSH-related diseases.Herein,we developed in vivo second near-infrared(NIR-II)window fluorescence(FL)and ratiometric photoacoustic(RPA)dual-modality imaging of GSH using a GSH-activatable probe(LET-14).LET-14 was synthesized based on a rhodamine hybrid xanthene skeleton with a FL shielding 2,4-dinitrobenzene sulfonyl group that can be specifically cleaved by GSH,thus resulting in a markedly bathochromic-shift absorption,a 6.5-fold increase in NIR-II FL intensity(FL920)and a 13-fold increase in RPA signal(PA880/PA705)in vitro.Intriguingly,LET-14 exhibits good selectivity and sensitivity for NIR-II FL and RPA dual-modality imaging of GSH in 4T1 tumor-bearing mouse model.Our findings develop an in vivo detection tool of GSH,which has great potential in the field of cancer diagnosis.展开更多
Vascular abnormalities are closely associated with the pathogenesis and progression of numerous diseases, such as thrombosis, tumors, and diabetes. Blood flow velocity serves as a critical biomarker for evaluating per...Vascular abnormalities are closely associated with the pathogenesis and progression of numerous diseases, such as thrombosis, tumors, and diabetes. Blood flow velocity serves as a critical biomarker for evaluating perfusion status. Quantitative detection of full-field blood flow variations in lesion areas holds significant scientific and clinical value for pathological studies,diagnosis, and intraoperative monitoring of related diseases. While laser speckle contrast imaging(LSCI) enables full-field blood flow visualization, its reliance on frame-based sensors necessitates handling massive data volumes, leading to inherent trade-offs among spatiotemporal resolution, real-time performance, and quantitative capabilities. Leveraging the asynchronous dynamic sensing, high temporal sampling rate, and low data redundancy of event cameras, this study proposes a quantitative blood flow imaging method termed laser speckle event imaging(LSEI). Experiments using off-the-shelf event cameras demonstrate that LSEI achieves real-time blood flow imaging with minimal computational overhead compared to frame-based LSCI. Furthermore,we investigate the relationship between event data streams and flow velocity through spatial-temporal autocorrelation analysis,enabling quantitative measurements without compromising temporal or spatial resolution. In in vivo imaging experiments of mouse ear blood flow, LSEI exhibits superior imaging details and real-time performance over conventional methods. The proposed approach holds promise as an efficient tool for diagnosis, therapeutic evaluation, and research on vascular-related diseases.展开更多
The photoacoustic imaging of lipid is intrinsically constrained by the feeble nature of endogenous lipid signals,posing a persistent sensitivity challenge that demands innovative solutions.Although adopting high-effic...The photoacoustic imaging of lipid is intrinsically constrained by the feeble nature of endogenous lipid signals,posing a persistent sensitivity challenge that demands innovative solutions.Although adopting high-efficiency excitation and detection elements may improve the imaging sensitivity to a certain extent,the application of the elements is inevitably subject to various limitations in practical applications,particularly during in vivo imaging and endoscopic imaging.In this study,we propose a multi-combinatorial approach to enhance the sensitivity of lipid photoacoustic imaging.The approach involves wavelet transform processing of one-dimensional A-line signals,gradient-based denoising of two-dimensional B-scan images,and finally,threedimensional spatial weighted averaging of the data processed by the previous two steps.This method not only significantly improves the signal-to-noise ratio(SNR)in distinguished feature regions of the image by around 10 dB,but also efficiently extracts weak signals with no distinct features in the original image.After processing with this method,the images acquired under single scanning were compared with those obtained under multiple scanning.The results showed highly consistent image features,with the structural similarity index increasing from 0.2 to 0.8,confirming the accuracy and reliability of the multi-combinatorial approach.展开更多
Generative Adversarial Networks(GANs)have become valuable tools in medical imaging,enabling realistic image synthesis for enhancement,augmentation,and restoration.However,their integration into clinical workflows rais...Generative Adversarial Networks(GANs)have become valuable tools in medical imaging,enabling realistic image synthesis for enhancement,augmentation,and restoration.However,their integration into clinical workflows raises concerns,particularly the risk of subtle distortions or hallucinations that may undermine diagnostic accuracy and weaken trust in AI-assisted decision-making.To address this challenge,we propose a hybrid deep learning framework designed to detect GAN-induced artifacts in medical images,thereby reinforcing the reliability of AI-driven diagnostics.The framework integrates low-level statistical descriptors,including high-frequency residuals and Gray-Level Co-occurrence Matrix(GLCM)texture features,with high-level semantic representations extracted from a pre-trained ResNet18.This dual-stream approach enables detection of both pixel-level anomalies and structural inconsistencies introduced by GAN-based manipulation.We validated the framework on a curated dataset of 10,000 medical images,evenly split between authentic and GAN-generated samples across four modalities:MRI,CT,X-ray,and fundus photography.To improve generalizability to real-world clinical settings,we incorporated domain adaptation strategies such as adversarial training and style transfer,reducing domain shift by 15%.Experimental results demonstrate robust performance,achieving 92.6%accuracy and an F1-score of 0.91 on synthetic test data,and maintaining strong performance on real-world GAN-modified images with 87.3%accuracy and an F1-score of 0.85.Additionally,the model attained an AUC of 0.96 and an average precision of 0.92,outperforming conventional GAN detection pipelines and baseline Convolutional Neural Network(CNN)architectures.These findings establish the proposed framework as an effective and reliable solution for detecting GAN-induced hallucinations in medical imaging,representing an important step toward building trustworthy and clinically deployable AI systems.展开更多
In recent years,the rapid advancement of artificial intelligence(AI)technology has enabled AI-assisted negative screening to significantly enhance physicians'efficiency through image feature analysis and multimoda...In recent years,the rapid advancement of artificial intelligence(AI)technology has enabled AI-assisted negative screening to significantly enhance physicians'efficiency through image feature analysis and multimodal data modeling,allowing them to focus more on diagnosing positive cases.Meanwhile,multispectral imaging(MSI)integrates spectral and spatial resolution to capture subtle tissue features invisible to the human eye,providing high-resolution data support for pathological analysis.Combining AI technology with MSI and employing quantitative methods to analyze multiband biomarkers(such as absorbance differences in keratin pearls)can effectively improve diagnostic specificity and reduce subjective errors in manual slide interpretation.To address the challenge of identifying negative tissue sections,we developed a discrimination algorithm powered by MSI.We demonstrated its efficacy using cutaneous squamous cell carcinoma(cSCC)as a representative case study.The algorithm achieved 100%accuracy in excluding negative cases and effectively mitigated the false-positive problem caused by cSCC heterogeneity.We constructed a multispectral image(MSI)dataset acquired at 520 nm,600 nm,and 630 nm wavelengths.Subsequently,we employed an optimized MobileViT model for tissue classification and performed comparative analyses against other models.The experimental results showed that our optimized MobileViT model achieved superior performance in identifying negative tissue sections,with a perfect accuracy rate of 100%.Thus,our results confirm the feasibility of integrating MSI with AI to exclude negative cases with perfect accuracy,offering a novel solution to alleviate the workload of pathologists.展开更多
Optical imaging has been pivotal in biological research(e.g.,cellular/developmental biology)for over two centuries.Recent advances like super-resolution fluorescence and nonlinear optical microscopy enable nanoscale s...Optical imaging has been pivotal in biological research(e.g.,cellular/developmental biology)for over two centuries.Recent advances like super-resolution fluorescence and nonlinear optical microscopy enable nanoscale studies of live cells and animals,yet their application to marine mollusks-key marine ecosystem species,remains underexplored.This review summarizes optical imaging techniques and their use in investigating marine mollusks across molecular,cellular,tissue,and individual levels.It highlights promising avenues for novel imaging methods to unravel the structures and functions of these organisms in future research,with a focus on advancements in applying cutting-edge optical techniques across these hierarchical levels.Given optical imaging's significance in elucidating marine mollusks'ecological and genetic information,this field deserves substantial attention and support.The review aims to address existing gaps,providing researchers and practitioners with comprehensive insights to foster further progress in this domain.展开更多
基金supported by the National Natural Science Foundation of China(Grant Nos.62001249 and62375140)。
文摘Color Fourier single-pixel imaging(FSI)enables efficient spectral and spatial imaging.Here,we propose a Fourier single-pixel imaging scheme with a random color filter array(FSI-RCFA).The proposed method employs a random color filter array(RCFA)to modulate Fourier patterns.A three-step phase-shifting technique reconstructs the Fourier spectrum,followed by an RCFA-based demosaicing algorithm to recover color images.Compared to traditional color FSI based on Bayer color filter array schemes(FSI-BCFA),our approach achieves superior separation between chrominance and luminance components in the frequency domain.Simulation results demonstrate that the FSI-RCFA method achieves a lower mean squared error(MSE),a higher peak signal-to-noise ratio(PSNR),and superior noise resistance compared to FSI-BCFA,while enabling direct single-channel pixel measurements for targeted applications such as agricultural defect detection.
基金upported by the National Natural Science Foundation of China(Grant No.62305184)the Major Key Project of Pengcheng Laboratory(Grant No.PCL2024A1)+1 种基金the Basic and Applied Basic Research Foundation of Guangdong Province(Grant No.2023A1515012932)the Science,Technology and Innovation Commission of Shenzhen Municipality(Grant No.WDZC20220818100259004).
文摘Single-pixel imaging(SPI)enables efficient sensing in challenging conditions.However,the requirement for numerous samplings constrains its practicality.We address the challenge of high-quality SPI reconstruction at ultra-low sampling rates.We develop an alternative optimization with physics and a data-driven diffusion network(APD-Net).It features alternative optimization driven by the learned task-agnostic natural image prior and the task-specific physics prior.During the training stage,APD-Net harnesses the power of diffusion models to capture data-driven statistics of natural signals.In the inference stage,the physics prior is introduced as corrective guidance to ensure consistency between the physics imaging model and the natural image probability distribution.Through alternative optimization,APD-Net reconstructs data-efficient,high-fidelity images that are statistically and physically compliant.To accelerate reconstruction,initializing images with the inverse SPI physical model reduces the need for reconstruction inference from 100 to 30 steps.Through both numerical simulations and real prototype experiments,APD-Net achieves high-quality,full-color reconstructions of complex natural images at a low sampling rate of 1%.In addition,APD-Net’s tuning-free nature ensures robustness across various imaging setups and sampling rates.Our research offers a broadly applicable approach for various applications,including but not limited to medical imaging and industrial inspection.
基金supported by the National Natural Science Foundation of China(Grant Nos.62322502,62131003,and 62088101)the Guangdong Province Key Laboratory of Intelligent Detection in Complex Environment of Aerospace,Land and Sea(Grant No.2022KSYS016).
文摘Existing single-pixel imaging(SPI)and sensing techniques suffer from poor reconstruction quality and heavy computation cost,limiting their widespread application.To tackle these challenges,we propose a large-scale single-pixel imaging and sensing(SPIS)technique that enables high-quality megapixel SPI and highly efficient image-free sensing with a low sampling rate.Specifically,we first scan and sample the entire scene using small-size optimized patterns to obtain information-coupled measurements.Compared with the conventional full-sized patterns,small-sized optimized patterns achieve higher imaging fidelity and sensing accuracy with 1 order of magnitude fewer pattern parameters.Next,the coupled measurements are processed through a transformer-based encoder to extract high-dimensional features,followed by a task-specific plugand-play decoder for imaging or image-free sensing.Considering that the regions with rich textures and edges are more difficult to reconstruct,we use an uncertainty-driven self-adaptive loss function to reinforce the network’s attention to these regions,thereby improving the imaging and sensing performance.Extensive experiments demonstrate that the reported technique achieves 24.13 dB megapixel SPI at a sampling rate of 3%within 1 s.In terms of sensing,it outperforms existing methods by 12%on image-free segmentation accuracy and achieves state-of-the-art image-free object detection accuracy with an order of magnitude less data bandwidth.
基金supported by the Natural Science Foundation of China Project(No.62525102).
文摘Single-pixel imaging(SPI)is a prominent scattering media imaging technique that allows image transmission via one-dimensional detection under structured illumination,with applications spanning from long-range imaging to microscopy.Recent advancements leveraging deep learning(DL)have significantly improved SPI performance,especially at low compression ratios.However,most DL-based SPI methods proposed so far rely heavily on extensive labeled datasets for supervised training,which are often impractical in real-world scenarios.Here,we propose an unsupervised learningenabled label-free SPI method for resilient information transmission through unknown dynamic scattering media.Additionally,we introduce a physics-informed autoencoder framework to optimize encoding schemes,further enhancing image quality at low compression ratios.Simulation and experimental results demonstrate that high-efficiency data transmission with structural similarity exceeding 0.9 is achieved through challenging turbulent channels.Moreover,experiments demonstrate that in a 5 m underwater dynamic turbulent channel,USAF target imaging quality surpasses traditional methods by over 13 dB.The compressive encoded transmission of 720×720 resolution video exceeding 30 seconds with great fidelity is also successfully demonstrated.These preliminary results suggest that our proposed method opens up a new paradigm for resilient information transmission through unknown dynamic scattering media and holds potential for broader applications within many other scattering media imaging technologies.
文摘In recent years,deep learning has been introduced into the field of Single-pixel imaging(SPI),garnering significant attention.However,conventional networks still exhibit limitations in preserving image details.To address this issue,we integrate Large Kernel Convolution(LKconv)into the U-Net framework,proposing an enhanced network structure named U-LKconv network,which significantly enhances the capability to recover image details even under low sampling conditions.
基金Project supported by the Natural Science Foundation of Hebei Province,China(Grant Nos.A2022201039 and F2019201446)the MultiYear Research Grant of University of Macao,China(Grant No.MYRG2020-00082-IAPME)+2 种基金the Science and Technology Development Fund from Macao SAR(FDCT),China(Grant No.0062/2020/AMJ)the Advanced Talents Incubation Program of the Hebei University(Grant No.8012605)the National Natural Science Foundation of China(Grant Nos.11204062,61774053,and 11674273)。
文摘We propose a method of complex-amplitude Fourier single-pixel imaging(CFSI)with coherent structured illumination to acquire both the amplitude and phase of an object.In the proposed method,an object is illustrated by a series of coherent structured light fields,which are generated by a phase-only spatial light modulator,the complex Fourier spectrum of the object can be acquired sequentially by a single-pixel photodetector.Then the desired complex-amplitude image can be retrieved directly by applying an inverse Fourier transform.We experimentally implemented this CFSI with several different types of objects.The experimental results show that the proposed method provides a promising complex-amplitude imaging approach with high quality and a stable configuration.Thus,it might find broad applications in optical metrology and biomedical science.
基金Project supported by the National Natural Science Foundation of China(Grant No.62075241).
文摘Single-pixel imaging(SPI)can transform 2D or 3D image data into 1D light signals,which offers promising prospects for image compression and transmission.However,during data communication these light signals in public channels will easily draw the attention of eavesdroppers.Here,we introduce an efficient encryption method for SPI data transmission that uses the 3D Arnold transformation to directly disrupt 1D single-pixel light signals and utilizes the elliptic curve encryption algorithm for key transmission.This encryption scheme immediately employs Hadamard patterns to illuminate the scene and then utilizes the 3D Arnold transformation to permutate the 1D light signal of single-pixel detection.Then the transformation parameters serve as the secret key,while the security of key exchange is guaranteed by an elliptic curve-based key exchange mechanism.Compared with existing encryption schemes,both computer simulations and optical experiments have been conducted to demonstrate that the proposed technique not only enhances the security of encryption but also eliminates the need for complicated pattern scrambling rules.Additionally,this approach solves the problem of secure key transmission,thus ensuring the security of information and the quality of the decrypted images.
基金Project supported by the National Key Research and Development Program of China (Grant No.2018YFB0504302)。
文摘We propose a single-pixel imaging(SPI)method to achieve a higher-resolution image via the Hadamard transform matrix.Unlike traditional SPI schemes,this new method recovers images by correlating single-pixel signals with synchronized transformed patterns of Hadamard bases that are actually projected onto the digital micromirror device.Each transform pattern is obtained through the inverse Fourier transform of the pattern acquired by Gaussian filtering of each Hadamard basis in the frequency domain.The proposed scheme is based on a typical SPI experimental setup and does not add any hardware complexity,enabling the transformation of Hadamard matrices and image reconstruction through data processing alone.Therefore,this approach could be considered as an alternative option for achieving fast SPI in a diffraction-limited imaging system,without the need for additional hardware.
基金supported by the National Natural Science Foundation of China(Grant No.61271376)the Anhui Provincial Natural Science Foundation,China(Grant No.1208085MF114)
文摘The single-pixel imaging(SPI) technique is able to capture two-dimensional(2 D) images without conventional array sensors by using a photodiode. As a novel scheme, Fourier single-pixel imaging(FSI) has been proven capable of reconstructing high-quality images. Due to the fact that the Fourier basis patterns(also known as grayscale sinusoidal patterns)cannot be well displayed on the digital micromirror device(DMD), a fast FSI system is proposed to solve this problem by binarizing Fourier pattern through a dithering algorithm. However, the traditional dithering algorithm leads to low quality as the extra noise is inevitably induced in the reconstructed images. In this paper, we report a better dithering algorithm to binarize Fourier pattern, which utilizes the Sierra–Lite kernel function by a serpentine scanning method. Numerical simulation and experiment demonstrate that the proposed algorithm is able to achieve higher quality under different sampling ratios.
文摘A two-stage training method is proposed to enhance imaging quality and reduce reconstruction time in datadriven single-pixel imaging(SPI)under undersampling conditions.This approach leverages a deep learning algorithm to simulate single-pixel detection and image reconstruction.During the initial training stage,an L2 regularization constraint is imposed on convolution modulation patterns to determine the optimal initial network weights.In the subsequent stage,a coupled deep learning method integrating coded-aperture design and SPI is adopted,which utilizes backpropagation of the loss function to iteratively optimize both the binarized modulation patterns and imaging network parameters.By reducing the binarization errors introduced by the dithering algorithm,this approach improves the quality of data-driven SPI.Compared with traditional deep-learning SPI methods,the proposed method significantly reduces computational complexity,resulting in accelerated image reconstruction.Experimental and simulation results demonstrate the advantages of the method,including high imaging quality,short image reconstruction time,and simplified training.For an image size of 64×64 pixels and 10%sampling rate,the proposed method achieves a peak signal-to-noise ratio of 23.22 dB,structural similarity index of 0.76,and image reconstruction time of approximately 2.57×10^(−4) seconds.
基金supported by the Shanxi Province Science and Technology Major Special Project(No.202201010101005)the Innovation Program for Quantum Science and Technology(No.2021ZD0300705)+6 种基金the National Natural Science Foundation of China(Nos.U23A20380,62105193,62127817,62075120,62075122,U22A2091,62222509,62205187,and62305200)the China Postdoctoral Science Foundation(No.2022M722006)the National Key Research and Development Program of China(No.2022YFA1404201)the Shanxi Province Science and Technology Innovation Talent Team(No.202204051001014)the Science and Technology Cooperation Project of Shanxi Province(No.202104041101021)the Shanxi“1331 Project”111 Projects(No.D18001)。
文摘Photon-level single-pixel imaging overcomes the reliance of traditional imaging techniques on large-scale array detectors,offering the advantages such as high sensitivity,high resolution,and efficient photon utilization.In this paper,we propose a photon-level dynamic feature single-pixel imaging method,leveraging the frequency domain sparsity of the object's dynamic features to construct a compressed measurement system through discrete random photon detection.In the experiments,we successfully captured 167 and 200 Hz featured frequencies and achieved high-quality image reconstruction with a data compression ratio of 20%.Our approach introduces a new detection dimension,significantly expanding the applications of photon-level single-pixel imaging in practical scenarios.
基金supported by the Innovation Project of Optics Valley Laboratory(No.OVL2023ZD005).
文摘Existing two-step Fourier single-pixel imaging (FSPI) suffers from low noise-robustness, and three-step FSPI and four-step FSPI improve the noise-robustness but at the cost of more measurements. In this Letter, we propose a method to improve the noise-robustness of two-step FSPI without additional illumination patterns or measurements. In the proposed method, the measurements from base patterns are replaced by the average values of the measurement from two sets of phase-shift patterns. Thus, the imaging degradation caused by the noise in the measurements from base patterns can be avoided, and more reliable Fourier spectral coefficients are obtained. The imaging quality of the proposed robust two-step FSPI is similar to those of three-step FSPI and four-step FSPI. Simulations and experimental results validate the effectiveness of the proposed method.
基金supported by the National Natural Science Foundation of China(Nos.12105143 and 11975121)the China Postdoctoral Science Foundation(No.2023M741453)+1 种基金the Engineering Research Center of Nuclear Technology Application(No.HJSJYB2020-1)the Key Laboratory of Ionizing Radiation Metering and Safety Evaluation for Jiangsu Province Market Regulation,and the Jiangsu Province Excellent Postdoctoral Program(No.JB23057).
文摘Neutron-induced gamma-ray imaging is a spectroscopic technique that uses characteristic gamma rays to infer the elemental distribution of an object.Currently,this technique requires the use of large facilities to supply a high neutron flux and a time-consuming detection procedure involving direct collimating measurements.In this study,a new method based on low neutron flux was proposed.A single-pixel gamma-ray detector combined with random pattern gamma-ray masks was used to measure the characteristic gamma rays emitted from the sample.Images of the elemental distribution in the sample,comprising 30×30 pixels,were reconstructed using the maximum-likelihood expectation-maximization algorithm.The results demonstrate that the elemental imaging of the sample can be accurately determined using this method.The proposed approach,which eliminates the need for high neutron flux and scanning measurements,can be used for in-field imaging applications.
文摘Background:Sarcomatoid carcinoma of the ureter(SCU)is a highly aggressive and relatively uncommon malignant tumor of the urinary tract.Its frequency is quite low,and its prognosis is very bad when compared to other cancers of the urinary system.SCU clinical reports are still hard to come by.MRI and PEI/CT imaging of ureteral sarcomatoid cancer is presented in this case to promote diagnostic awareness and comprehension of the imaging characteristics of this uncommon illness.Method:The patient had ureteral sarcomatoid cancer,which was verified by pathological investigation after ureteroscopic biopsy.The patient’s clinical information,imaging results,surgical outcomes,and pathological findings were gathered.A retrospective study was carried out in combinationwith pertinent national and international literature.Results:An 84-year-old female patient was admitted for“left flank discomfort lasting over one month.”MRI revealed an irregular soft tissue mass in the middle-lower segment of the left ureter.T2-weighted imaging showed an unevenly slightly hyperintense signal.Diffusion-weighted imaging demonstrated restricted diffusion.Contrastenhanced imaging exhibited heterogeneous enhancement.PET/CT demonstrated significantly increased fluorodeoxyglucose metabolism in the mass with secondary left upper urinary tract obstruction.Concurrent findings included a solitary metastatic lesion in hepatic segment S6 and multiple lymph node metastases along the left common iliac and external iliac arteries.Preoperative diagnosis suggested a malignant tumor of the ureter.The patient underwent left nephroureteroscopy with biopsy,and the postoperative pathological diagnosis was ureteral sarcomatoid carcinoma.Conclusion:Ureteral sarcomatoid carcinoma is a rare,highly malignant,and aggressive tumor with nonspecific imaging features,typically presenting as an invasively growing mass.Diagnosis relies on postoperative pathology and immunohistochemical examination.MRI and PET/CT scans are valuable for preoperative localization and characterization,tumor staging,treatment planning,and postoperative follow-up.The prognosis is extremely negative.The main treatment option is radical surgery,although constant monitoring is necessary since early recurrence and metastases are frequent after surgery.
基金supported by the National Natural Science Foundation of China(Nos.82372116,U23A2097)Guangdong Basic and Applied Basic Research Foundation(No.2022A1515010620)+2 种基金Shenzhen Medical Research Fund(Nos.B2302047,A2302047)Shenzhen Science and Technology Program(No.JCYJ20220818095806014)Research Team Cultivation Program of Shenzhen University(No.2023QNT019).
文摘The level of glutathione(GSH)is significantly associated with numerous pathological processes,thus,real-time detection of the GSH level is of significance for early diagnosis of GSH-related diseases.Herein,we developed in vivo second near-infrared(NIR-II)window fluorescence(FL)and ratiometric photoacoustic(RPA)dual-modality imaging of GSH using a GSH-activatable probe(LET-14).LET-14 was synthesized based on a rhodamine hybrid xanthene skeleton with a FL shielding 2,4-dinitrobenzene sulfonyl group that can be specifically cleaved by GSH,thus resulting in a markedly bathochromic-shift absorption,a 6.5-fold increase in NIR-II FL intensity(FL920)and a 13-fold increase in RPA signal(PA880/PA705)in vitro.Intriguingly,LET-14 exhibits good selectivity and sensitivity for NIR-II FL and RPA dual-modality imaging of GSH in 4T1 tumor-bearing mouse model.Our findings develop an in vivo detection tool of GSH,which has great potential in the field of cancer diagnosis.
基金supported by the National Natural Science Foundation of China (Grant No.12572210)the Scientific Instrument Developing Project of Shenzhen University (Grant Nos.2023YQ011,2024YQ001)the Shenzhen Science and Technology Innovation Commission Project—Stable Support (General Project)(Grant No.20231120175055001)。
文摘Vascular abnormalities are closely associated with the pathogenesis and progression of numerous diseases, such as thrombosis, tumors, and diabetes. Blood flow velocity serves as a critical biomarker for evaluating perfusion status. Quantitative detection of full-field blood flow variations in lesion areas holds significant scientific and clinical value for pathological studies,diagnosis, and intraoperative monitoring of related diseases. While laser speckle contrast imaging(LSCI) enables full-field blood flow visualization, its reliance on frame-based sensors necessitates handling massive data volumes, leading to inherent trade-offs among spatiotemporal resolution, real-time performance, and quantitative capabilities. Leveraging the asynchronous dynamic sensing, high temporal sampling rate, and low data redundancy of event cameras, this study proposes a quantitative blood flow imaging method termed laser speckle event imaging(LSEI). Experiments using off-the-shelf event cameras demonstrate that LSEI achieves real-time blood flow imaging with minimal computational overhead compared to frame-based LSCI. Furthermore,we investigate the relationship between event data streams and flow velocity through spatial-temporal autocorrelation analysis,enabling quantitative measurements without compromising temporal or spatial resolution. In in vivo imaging experiments of mouse ear blood flow, LSEI exhibits superior imaging details and real-time performance over conventional methods. The proposed approach holds promise as an efficient tool for diagnosis, therapeutic evaluation, and research on vascular-related diseases.
基金supported by the National Key Research and Development Program of China(2022YFC2402400)the National Natural Science Foundation of China(82027803,62275062)+7 种基金the Guangdong Provincial Key Laboratory of Biomedical Optical Imaging Technology(2020B121201010)the Shenzhen Science and Technology Innovation Committee under Grant(JCYJ20220818101417039)the Shenzhen Key Laboratory for Molecular lmaging(ZDSY20130401165820357)the Shenzhen Medical Research Fund(D2404002)the Project of Shandong Innovation and Startup Community of High-end Medical Apparatus and Instruments(2023-SGTTXM-002 and 2024-SGTTXM-005)the Shandong Province Technology Innovation Guidance Plan(Central Leading Local Science and Technology Development Fund)(YDZX2023115)the Taishan Scholar Special Funding Project of Shandong Provinceand the Shandong Laboratory of Advanced Biomaterials and Medical Devices in Weihai(ZL202402).
文摘The photoacoustic imaging of lipid is intrinsically constrained by the feeble nature of endogenous lipid signals,posing a persistent sensitivity challenge that demands innovative solutions.Although adopting high-efficiency excitation and detection elements may improve the imaging sensitivity to a certain extent,the application of the elements is inevitably subject to various limitations in practical applications,particularly during in vivo imaging and endoscopic imaging.In this study,we propose a multi-combinatorial approach to enhance the sensitivity of lipid photoacoustic imaging.The approach involves wavelet transform processing of one-dimensional A-line signals,gradient-based denoising of two-dimensional B-scan images,and finally,threedimensional spatial weighted averaging of the data processed by the previous two steps.This method not only significantly improves the signal-to-noise ratio(SNR)in distinguished feature regions of the image by around 10 dB,but also efficiently extracts weak signals with no distinct features in the original image.After processing with this method,the images acquired under single scanning were compared with those obtained under multiple scanning.The results showed highly consistent image features,with the structural similarity index increasing from 0.2 to 0.8,confirming the accuracy and reliability of the multi-combinatorial approach.
基金supported and funded by the Deanship of Scientific Research at Imam Mohammad Ibn Saud Islamic University(IMSIU)(grant number IMSIU-DDRSP2601).
文摘Generative Adversarial Networks(GANs)have become valuable tools in medical imaging,enabling realistic image synthesis for enhancement,augmentation,and restoration.However,their integration into clinical workflows raises concerns,particularly the risk of subtle distortions or hallucinations that may undermine diagnostic accuracy and weaken trust in AI-assisted decision-making.To address this challenge,we propose a hybrid deep learning framework designed to detect GAN-induced artifacts in medical images,thereby reinforcing the reliability of AI-driven diagnostics.The framework integrates low-level statistical descriptors,including high-frequency residuals and Gray-Level Co-occurrence Matrix(GLCM)texture features,with high-level semantic representations extracted from a pre-trained ResNet18.This dual-stream approach enables detection of both pixel-level anomalies and structural inconsistencies introduced by GAN-based manipulation.We validated the framework on a curated dataset of 10,000 medical images,evenly split between authentic and GAN-generated samples across four modalities:MRI,CT,X-ray,and fundus photography.To improve generalizability to real-world clinical settings,we incorporated domain adaptation strategies such as adversarial training and style transfer,reducing domain shift by 15%.Experimental results demonstrate robust performance,achieving 92.6%accuracy and an F1-score of 0.91 on synthetic test data,and maintaining strong performance on real-world GAN-modified images with 87.3%accuracy and an F1-score of 0.85.Additionally,the model attained an AUC of 0.96 and an average precision of 0.92,outperforming conventional GAN detection pipelines and baseline Convolutional Neural Network(CNN)architectures.These findings establish the proposed framework as an effective and reliable solution for detecting GAN-induced hallucinations in medical imaging,representing an important step toward building trustworthy and clinically deployable AI systems.
基金funded by the Natural Science Foundation of Shanghai Municipality(No.21ZR1440500)the Shanghai Science and Technology Commission(Grant No.21S31902700).
文摘In recent years,the rapid advancement of artificial intelligence(AI)technology has enabled AI-assisted negative screening to significantly enhance physicians'efficiency through image feature analysis and multimodal data modeling,allowing them to focus more on diagnosing positive cases.Meanwhile,multispectral imaging(MSI)integrates spectral and spatial resolution to capture subtle tissue features invisible to the human eye,providing high-resolution data support for pathological analysis.Combining AI technology with MSI and employing quantitative methods to analyze multiband biomarkers(such as absorbance differences in keratin pearls)can effectively improve diagnostic specificity and reduce subjective errors in manual slide interpretation.To address the challenge of identifying negative tissue sections,we developed a discrimination algorithm powered by MSI.We demonstrated its efficacy using cutaneous squamous cell carcinoma(cSCC)as a representative case study.The algorithm achieved 100%accuracy in excluding negative cases and effectively mitigated the false-positive problem caused by cSCC heterogeneity.We constructed a multispectral image(MSI)dataset acquired at 520 nm,600 nm,and 630 nm wavelengths.Subsequently,we employed an optimized MobileViT model for tissue classification and performed comparative analyses against other models.The experimental results showed that our optimized MobileViT model achieved superior performance in identifying negative tissue sections,with a perfect accuracy rate of 100%.Thus,our results confirm the feasibility of integrating MSI with AI to exclude negative cases with perfect accuracy,offering a novel solution to alleviate the workload of pathologists.
基金supported by the National Natural Science Foundation of China(T2421003/22327802/41806142)Guangdong Basic and Applied Basic Foundation(2025A1515011484/2022A1515011845)+1 种基金Shenzhen Key Laboratory of Photonics and Biophotonics(ZDSYS20210623092006020)Medical-Engineering Interdisciplinary Research Foundation of Shenzhen University(2023YG033).
文摘Optical imaging has been pivotal in biological research(e.g.,cellular/developmental biology)for over two centuries.Recent advances like super-resolution fluorescence and nonlinear optical microscopy enable nanoscale studies of live cells and animals,yet their application to marine mollusks-key marine ecosystem species,remains underexplored.This review summarizes optical imaging techniques and their use in investigating marine mollusks across molecular,cellular,tissue,and individual levels.It highlights promising avenues for novel imaging methods to unravel the structures and functions of these organisms in future research,with a focus on advancements in applying cutting-edge optical techniques across these hierarchical levels.Given optical imaging's significance in elucidating marine mollusks'ecological and genetic information,this field deserves substantial attention and support.The review aims to address existing gaps,providing researchers and practitioners with comprehensive insights to foster further progress in this domain.
