Search

Optical encryption is a potential scheme for information security that exploits abundant degrees of freedom of light to encode information. However, conventional encryption based on fluorescent materials faces challenges in handling complex secret information. Alternatively, single‐pixel imaging (SPI) provides a computational modality to solve these problems. In this study, a high‐capacity fluorescence encryption scheme, achieved by introducing lanthanide materials and steganography into the encoding and decoding processes of SPI is proposed. Two types of well‐designed lanthanide luminescent materials are utilized and excited to generate fluorescence images (fluo‐images), which are crucial in this scheme. Various practical experiments using fluo‐images as secret keys demonstrate the robustness, effectiveness, and repeatability of this scheme. Furthermore, multi‐image experiments indicate the potential of this method to increase secret information capacity. Thus, the proposed fluorescence encryption scheme does provide an efficient computational encryption strategy based on lanthanide luminescent materials for information security, which can improve the security of traditional optical encryption and simultaneously enhance the flexibility of SPI computational decryption. [ABSTRACT FROM AUTHOR]

With the advancement of computational imaging, a large number of spectral imaging systems based on encoding–decoding have emerged, among which phase‐encoding spectral imaging systems have attracted widespread interest. Conventional phase‐encoding systems suffer from severe image degradation and limited light throughput. To address these challenges and achieve video‐rate spectral imaging with high spatial resolution and spectral accuracy, a novel optical system based on diffractive‐refractive hybrid optics is proposed. Here, a diffractive optical element is employed to perform imaging and dispersion functions, while a rear lens is used to shorten the system's back focal length and reduce the size of the point spread function. Meanwhile, convolutional neural network‐based spectral reconstruction algorithms are employed to reconstruct the spectral data cubes from diffraction blurred images. A compact, cost‐effective, and portable prototype has been constructed, demonstrating the capability to acquire and reconstruct 30 spectral data cubes per second, each with dimensions of 1080×1280×43${1080 \times 1280 \times 43}$ in the spectral range of 480–900 nm with a 10 nm spectral interval. The optical system has the potential to broaden the application scope of phase‐encoding spectral imaging systems in various scenarios. [ABSTRACT FROM AUTHOR]

Learned optics, which incorporate lightweight diffractive optics, coded-aperture modulation, and specialized image-processing neural networks, have recently garnered attention in the field of snapshot hyperspectral imaging (HSI). While conventional methods typically rely on a single lens element paired with an off-the-shelf color sensor, these setups, despite their widespread availability, present inherent limitations. First, the Bayer sensor's spectral response curves are not optimized for HSI applications, limiting spectral fidelity of the reconstruction. Second, single lens designs rely on a single diffractive optical element (DOE) to simultaneously encode spectral information and maintain spatial resolution across all wavelengths, which constrains spectral encoding capabilities. This work investigates a multi-channel lens array combined with aperture-wise color filters, all co-optimized alongside an image reconstruction network. This configuration enables independent spatial encoding and spectral response for each channel, improving optical encoding across both spatial and spectral dimensions. Specifically, we validate that the method achieves over a 5dB improvement in PSNR for spectral reconstruction compared to existing single-diffractive lens and coded-aperture techniques. Experimental validation further confirmed that the method is capable of recovering up to 31 spectral bands within the 429--700 nm range in diverse indoor and outdoor environments. [ABSTRACT FROM AUTHOR]

Pixel binning-based Quad sensors (mega-pixel resolution camera sensor) offer a promising solution to address the hardware limitations of compact cameras for low-light imaging. However, the binning process leads to reduced spatial resolution and introduces non-Bayer CFA artifacts. In this paper, we propose a Quad CFA-driven remosaicing model that effectively converts noisy Quad Bayer and standard Bayer patterns compatible to existing Image Signal Processor (ISP) without any loss in resolution. To enhance the practicality of the remosaicing model for real-world images affected by mixed noise, we introduce a novel dual-head joint remosaicing and denoising network (DJRD), which addresses the order of denoising and remosaicing by performing them in parallel. In DJRD, we customize two denoising branches for Quad Bayer and Bayer inputs. These branches model non-local and local dependencies, CFA location, and frequency information using residual convolutional layers, Swin Transformer, and wavelet transform-based CNN. Furthermore, to improve the model's performance on challenging cases, we fine-tune DJRD to handle difficult scenarios by identifying problematic patches through Moire and zipper detection metrics. This post-training phase allows the model to focus on resolving complex image regions. Extensive experiments conducted on simulated and real images in both Bayer and sRGB domains demonstrate that DJRD outperforms competing models by approximately 3 dB, while maintaining the simplicity of implementation without adding any hardware. [ABSTRACT FROM AUTHOR]

Fourier ptychographic microscopy (FPM) breaks through the resolution limitations of conventional optical systems, which offer a full‐field view and high resolution without additional mechanical scanning. However, conventional image‐domain optimizations require trade‐offs between correction efficacy, data redundancy, and reconstruction accuracy. Furthermore, the existing linear time‐invariant model for actual nonlinear, time‐varying optical systems leads to forward model mismatch, complicating the corrections of the vignetting effect. To overcome these challenges and achieve stitching‐free FPM, a family of forward wavelet‐transform models (WL‐FPM) is proposed. WL‐FPM employs the reversibility of the wavelet transform for high‐fidelity reconstruction in the multiscale feature domain. The wavelet loss function is updated in each iteration, and non‐convex optimization is solved by complex back diffraction. WL‐FPM offers stitching‐free, high‐resolution, and robust reconstruction under various challenging conditions, including vignetting effects, LED position mismatch, intensity fluctuations, and high‐level noise environments, which outperform conventional FPM methods. Under a 4X objective with NA 0.1, WL‐FPM achieves a 435‐nm resolution and stitching‐free full‐field reconstruction of a 3.328 × 3.328 mm2 pathological section with distinct subcellular organelles. In live cell imaging, it provides a full‐field observation with distinct lipids in a single cell. A large number of simulation and experimental results demonstrate its potential for biomedical applications. [ABSTRACT FROM AUTHOR]

Super‐resolution structured‐illumination microscopy (SIM) is a powerful technique that allows one to surpass the diffraction limit by up to a factor two. Yet, its practical use is hampered by its sensitivity to imaging conditions which makes it prone to reconstruction artefacts. In this work, we present FlexSIM, a flexible SIM reconstruction method capable to handle highly challenging data. Specifically, we demonstrate the ability of FlexSIM to deal with the distortion of patterns, the high level of noise encountered in live imaging, as well as out‐of‐focus fluorescence. Moreover, we show that FlexSIM achieves state‐of‐the‐art performance over a variety of open SIM datasets. [ABSTRACT FROM AUTHOR]

Video snapshot compressive imaging (SCI) uses a low-speed 2D detector to capture high-speed scene, where the dynamic scene is modulated by different masks and then compressed into a snapshot measurement. Following this, a reconstruction algorithm is needed to reconstruct the high-speed video frames. Although state-of-the-art (SOTA) deep learning-based reconstruction algorithms have achieved impressive results, they still face the following challenges due to excessive model complexity and GPU memory limitations: (1) These models need high computational cost, and (2) They are usually unable to reconstruct large-scale video frames at high compression ratios. To address these issues, we develop an efficient network for video SCI by using hierarchical residual-like connections and hybrid CNN-Transformer structure within a single residual block, dubbed EfficientSCI++. The EfficientSCI++ network can well explore spatial-temporal correlation using convolution in the spatial domain and Transformer in the temporal domain, respectively. We are the first time to demonstrate that a UHD color video ( 1644 × 3840 × 3 ) with high compression ratio (40) can be reconstructed from a snapshot 2D measurement using a single end-to-end deep learning model with PSNR above 34 dB. Moreover, a mixed-precision model is trained to further accelerate the video SCI reconstruction process and save memory footprint. Extensive results on both simulation and real data demonstrate that, compared with precious SOTA methods, our proposed EfficientSCI++ and EfficientSCI can achieve comparable reconstruction quality with much cheaper computational cost and better real-time performance. Code is available at https://github.com/mcao92/EfficientSCI-plus-plus. [ABSTRACT FROM AUTHOR]

Advances in imaging tools have always been a pivotal driver for new discoveries in neuroscience. An ability to visualize neurons and subcellular structures deep within the brain of a freely behaving animal is integral to our understanding of the relationship between neural activity and higher cognitive functions. However, fast highresolution imaging is limited to sub-surface brain regions and generally requires head fixation of the animal under the microscope. Developing new approaches to address these challenges is critical. The last decades have seen rapid progress in minimally invasive endo-microscopy techniques based on bare optical fibers. A single multimode fiber can be used to penetrate deep into the brain without causing significant damage to the overlying structures and provide high-resolution imaging. Here, we discuss how the full potential of high-speed super-resolution fiber endoscopy can be realized by a holistic approach that combines fiber optics, light shaping, and advanced computational algorithms. The recent progress opens up new avenues for minimally invasive deep brain studies in freely behaving mice. [ABSTRACT FROM AUTHOR]

The intricate nature of the brain necessitates the application of advanced probing techniques to comprehensively study and understand its working mechanisms. Neurophotonics offers minimally invasive methods to probe the brain using optics at cellular and even molecular levels. However, multiple challenges persist, especially concerning imaging depth, field of view, speed, and biocompatibility. A major hindrance to solving these challenges in optics is the scattering nature of the brain. This perspective highlights the potential of complex media optics, a specialized area of study focused on light propagation in materials with intricate heterogeneous optical properties, in advancing and improving neuronal readouts for structural imaging and optical recordings of neuronal activity. Key strategies include wavefront shaping techniques and computational imaging and sensing techniques that exploit scattering properties for enhanced performance. We discuss the potential merger of the two fields as well as potential challenges and perspectives toward longer term in vivo applications. [ABSTRACT FROM AUTHOR]

Fourier ptychographic microscopy (FPM) is a computational imaging technology that can acquire high-resolution large-area images for applications ranging from biology to microelectronics. In this study, we utilize multifocal plane imaging to enhance the existing FPM technology. Using an RGB light emitting diode (LED) array to illuminate the sample, raw images are captured using a color camera. Then, exploiting the basic optical principle of wavelength-dependent focal length variation, three focal plane images are extracted from the raw image through simple R, G, and B channel separation. Herein, a single aspherical lens with a numerical aperture (NA) of 0.15 was used as the objective lens, and the illumination NA used for FPM image reconstruction was 0.08. Therefore, simultaneous multifocal plane FPM with a synthetic NA of 0.23 was achieved. The multifocal imaging performance of the enhanced FPM system was then evaluated by inspecting a transparent organic light-emitting diode (OLED) sample. The FPM system was able to simultaneously inspect the individual OLED pixels as well as the surface of the encapsulating glass substrate by separating R, G, and B channel images from the raw image, which was taken in one shot. [ABSTRACT FROM AUTHOR]

Digital images are corrupted with noise, and image denoising is an important step in image processing modules. In this review, the latest developments in filtering methods for color image restoration are analyzed. These algorithms are compared in terms of objective image quality measures and divided into major classes, such as spatial domain, switching and wavelet filtering methods. These classes are based on the particular methodology used in image denoising algorithms and further subdivided to show their classification in terms of noise models utilized, application style, and stages the filters applied in images. In particular, we present a review of filtering methods in color image denoising, published over the past two decades. Our classification and succinct descriptions of color image restoration by these mathematical filtering techniques and their characterizations can help choose the appropriate ones for various downstream image processing tasks. [ABSTRACT FROM AUTHOR]

Magnetic induction tomography (MIT) is an emerging imaging technology holding significant promise in the field of cerebral hemorrhage monitoring. The commonly employed imaging method in MIT is time-difference imaging. However, this approach relies on magnetic field signals preceding cerebral hemorrhage, which are often challenging to obtain. Multiple bioelectrical impedance information with different frequencies is added to this study on the basis of single-frequency information, and the collected signals with different frequencies are identified to obtain the magnetic field signal generated by single-layer heterogeneous tissue. The Stacked Autoencoder (SAE) neural network algorithm is used to reconstruct the images of head multi-layer tissues. Both numerical simulation and phantom experiments are carried out. The results indicate that the relative error of the multi-frequency SAE reconstruction is only 7. 82°/o, outperforming traditional algorithms. Moreover, under a noise level of 40 dB, the anti-interference capability of the MIT algorithm based on frequency identification and SAE is superior to traditional algorithms. This research explores a novel approach for the dynamic monitoring of cerebral hemorrhage and demonstrates the potential advantages of MIT in non-invasive monitoring. [ABSTRACT FROM AUTHOR]

Abstract We present a large-scale computational 3D topographic microscope that enables 6-gigapixel profilometric 3D imaging at micron-scale resolution across >110 cm2 areas over multi-millimeter axial ranges. Our computational microscope, termed STARCAM (Scanning Topographic All-in-focus Reconstruction with a Computational Array Microscope), features a parallelized, 54-camera architecture with 3-axis translation to capture, for each sample of interest, a multi-dimensional, 2.1-terabyte (TB) dataset, consisting of a total of 224,640 9.4-megapixel images. We developed a self-supervised neural network-based algorithm for 3D reconstruction and stitching that jointly estimates an all-in-focus photometric composite and 3D height map across the entire field of view, using multi-view stereo information and image sharpness as a focal metric. The memory-efficient, compressed differentiable representation offered by the neural network effectively enables joint participation of the entire multi-TB dataset during the reconstruction process. Validation experiments on gauge blocks demonstrate a profilometric precision and accuracy of 10 µm or better. To demonstrate the broad utility of our new computational microscope, we applied STARCAM to a variety of decimeter-scale objects, with applications ranging from cultural heritage to industrial inspection.

Traditional staining of biological specimens for microscopic imaging entails time-consuming, laborious, and costly procedures, in addition to producing inconsistent labeling and causing irreversible sample damage. In recent years, computational virtual staining using deep learning techniques has evolved into a robust and comprehensive application for streamlining the staining process without typical histochemical staining-related drawbacks. Such virtual staining techniques can also be combined with neural networks designed to correct various microscopy aberrations, such as out-of-focus or motion blur artifacts, and improve upon diffracted-limited resolution. Here, we highlight how such methods lead to a host of new opportunities that can significantly improve both sample preparation and imaging in biomedical microscopy.

Imaging is a longstanding research topic in optics and photonics and is an important tool for a wide range of scientific and engineering fields. Computational imaging is a powerful framework for designing innovative imaging systems by incorporating signal processing into optics. Conventional approaches involve individually designed optical and signal processing systems, which unnecessarily increased costs. Computational imaging, on the other hand, enhances the imaging performance of optical systems, visualizes invisible targets, and minimizes optical hardware. Digital holography and computer-generated holography are the roots of this field. Recent advances in information science, such as deep learning, and increasing computational power have rapidly driven computational imaging and have resulted in the reinvention these imaging technologies. In this paper, I survey recent research topics in computational imaging, where optical randomness is key. Imaging through scattering media, non-interferometric quantitative phase imaging, and real-time computer-generated holography are representative examples. These recent optical sensing and control technologies will serve as the foundations of next-generation imaging systems in various fields, such as biomedicine, security, and astronomy. [ABSTRACT FROM AUTHOR]

Compressed sampling allows to accurately reconstruct a sparse signal even in case of incomplete signal measurements. In this paper, we apply this method to single-pixel imaging and explore the possibilities of image reconstruction by sampling it with an incomplete set of binary light patterns. Using computer simulation, we optimize the image sampling process and find parameters of light patterns such that single-pixel imaging works best. [ABSTRACT FROM AUTHOR]

Single-pixel imaging allows to obtain images without the use of photosensors with spatial resolution. In this method, an image is calculated by measuring the image conformity to a given set of light patterns by a single-pixel detector. However, when implementing single-pixel imaging in practice, one has to deal with various imperfections, which lead to the difference between the experiment and the idealized theoretical model. In this work, we analyze the effect of detector noise on the ability to compute an image using a compressed sampling algorithm. By conducting computer simulations of single-pixel imaging, we investigate methods for suppressing the effects of detector noise and find optimum parameters of the measurement process. As a result, we demonstrate the ability to obtain images with a realistic model of the detector noise. [ABSTRACT FROM AUTHOR]

Large-aperture, lightweight, and high-resolution imaging are hallmarks of major optical systems. To eliminate aberrations, traditional systems are often bulky and complex, whereas the small volume and light weight of diffractive lenses position them as potential substitutes. However, their inherent diffraction mechanism leads to severe dispersion, which limits their application in wide spectral bands. Addressing the dispersion issue in diffractive lenses, we propose a chromatic aberration correction algorithm based on compressed sensing. Utilizing the diffractive lens's focusing ability at the reference wavelength and its degradation performance at other wavelengths, we employ compressed sensing to reconstruct images from incomplete image information. In this work, we design a harmonic diffractive lens with a diffractive order of M = 150 , an aperture of 40 mm, a focal length f 0 = 320 mm, a reference wavelength λ 0 = 550 nm, a wavelength range of 500–800 nm, and 7 annular zones. Through algorithmic recovery, we achieve clear imaging in the visible spectrum, with a peak signal-to-noise ratio (PSNR) of 22.85 dB, a correlation coefficient of 0.9596, and a root mean square error (RMSE) of 0.02, verifying the algorithm's effectiveness. [ABSTRACT FROM AUTHOR]

We present an efficient construction method for object recognition based on speckle learning using the optical memory effect. An object classifier based on speckle learning without the process of reducing or eliminating scattering and with a simple optical setup has been previously reported, but it requires a large number of training images to improve the performance of the classifier. This method is not applicable for bioimaging because of the difficulty of collecting training images caused by position control and phototoxicity of target cells. In our method, a wide variety of training images are augmented by a computer from a few speckle intensity images in the working range of the optical memory effect. We experimentally demonstrated our method with a 4f-optical system implementing the optical memory effect. As a result, the constructed binary classifier showed high accuracy under various scattering conditions and resolutions of the test image. [ABSTRACT FROM AUTHOR]

In recent years, metasurface, as a representative of micro- and nano-optics, have demonstrated a powerful ability to manipulate light, which can modulate a variety of physical parameters, such as wavelength, phase, and amplitude, to achieve various functions and substantially improve the performance of conventional optical components and systems. Artificial Intelligence (AI) is an emerging strong and effective computational tool that has been rapidly integrated into the study of physical sciences over the decades and has played an important role in the study of metasurface. This review starts with a brief introduction to the basics and then describes cases where AI and metasurface research have converged: from AI-assisted design of metasurface elements up to advanced optical systems based on metasurface. We demonstrate the advanced computational power of AI, as well as its ability to extract and analyze a wide range of optical information, and analyze the limitations of the available research resources. Finally conclude by presenting the challenges posed by the convergence of disciplines. [ABSTRACT FROM AUTHOR]

Image reconstruction is a crucial aspect of computational imaging. The compressed sensing reconstruction (CS) method has been developed to obtain high-quality images. However, the CS method is commonly time-consuming in image reconstruction. To overcome this drawback, we propose a compressed sensing reconstruction method with fast convolution filtering (F-CS method), which significantly increases reconstruction speed by reducing the number of convolution operations without image fill. The experimental results show that by using the F-CS method, the reconstruction speed can be increased by a factor of 7 compared to the conventional CS method. Moreover, the F-CS method proposed in this paper is compared with the back-propagation reconstruction (BP) method and super-resolution reconstruction (SR) method, and it is validated that the proposed method has a lower computational resource cost for high-quality image reconstruction and exhibits a much more balanced capability. [ABSTRACT FROM AUTHOR]

An early identification and subsequent management of cerebral small vessel disease (cSVD) grade 1 can delay progression into grades II and III. Machine learning algorithms have shown considerable promise in medical image interpretation automation. An experimental cross-sectional study aimed to develop an automated computer-aided diagnostic system based on AI (artificial intelligence) tools to detect grade 1-cSVD with improved accuracy. Patients with Fazekas grade 1 cSVD on Non-Contrast Magnetic Resonance Imaging (MRI) Brain of age >40 years of both genders were included. The dataset was pre-processed to be fed into a 3D convolutional neural network (CNN) model. A 3D stack with the shape (120, 128, 128, 1) containing axial slices from the brain magnetic resonance image was created. The model was created from scratch and contained four convolutional and three fully connected (FC) layers. The dataset was preprocessed by making a 3D stack, and normalizing, resizing, and completing the stack was performed. A 3D-CNN model architecture was designed to train and test preprocessed images. We achieved an accuracy of 93.12 % when 2D axial slices were used. When the 2D slices of a patient were stacked to form a 3D image, an accuracy of 85.71 % was achieved on the test set. Overall, the 3D-CNN model performed very well on the test set. The earliest and the most accurate diagnosis from computational imaging methods can help reduce the huge burden of cSVD and its associated morbidity in the form of vascular dementia.

Quantitative susceptibility mapping (QSM) is a rising MRI-based technology and quite a few QSM-related algorithms have been proposed to reconstruct maps of tissue susceptibility distribution from phase images. In this paper, we develop a comprehensive susceptibility imaging process and analysis studio (SIPAS) that can accomplish reliable QSM processing and offer a standardized evaluation system. Specifically, SIPAS integrates multiple methods for each step, enabling users to select algorithm combinations according to data conditions, and QSM maps could be evaluated by two aspects, including image quality indicators within all voxels and region-of-interest (ROI) analysis. Through a sophisticated design of user-friendly interfaces, the results of each procedure are able to be exhibited in axial, coronal, and sagittal views in real-time, meanwhile ROIs can be displayed in 3D rendering visualization. The accuracy and compatibility of SIPAS are demonstrated by experiments on multiple in vivo human brain datasets acquired from 3T, 5T, and 7T MRI scanners of different manufacturers. We also validate the QSM maps obtained by various algorithm combinations in SIPAS, among which the combination of iRSHARP and SFCR achieves the best results on its evaluation system. SIPAS is a comprehensive, sophisticated, and reliable toolkit that may prompt the QSM application in scientific research and clinical practice.

In the semiconductor industry, the demand for more precise and accurate overlay metrology tools has increased because of the continued shrinking of feature sizes in integrated circuits. To achieve the required sub-nanometre precision, the current technology for overlay metrology has become complex and is reaching its limits. Herein, we present a dark-field digital holographic microscope using a simple two-element imaging lens with a high numerical aperture capable of imaging from the visible to near-infrared regions. This combination of high resolution and wavelength coverage was achieved by combining a simple imaging lens with a fast and accurate correction of non-isoplanatic aberrations. We present experimental results for overlay targets that demonstrate the capability of our computational aberration correction in the visible and near-infrared wavelength regimes. This wide-ranged-wavelength imaging system can advance semiconductor metrology.

Conventional iterative phase retrieval suffers from an inherent phase ambiguity due to limited measurement intensity. Multimodal amplitude modulation introduces physical constraints to tackle the underdetermination challenge. However, the time overhead caused by mask switching slows down the imaging speed. To increase imaging speed, we report an accelerated coded phase retrieval method by optimizing modulation masks. Compared to existing methods that require at least four patterns as inputs, the proposed method requires only three mask modulations to robustly reconstruct complex objects. The transparent pixels of the two masks partially overlap, constituting a strong constraint on the objective function. An additional random mask increases the difference between diffraction intensity patterns and ensures that the algorithm converges. The proposed method of efficient modulation using pure amplitude elements may open the door to short-wavelength high-speed complex amplitude imaging. Numerical simulations and proof-of-principle experiments have verified the feasibility of this method.

Diffractive optical imaging spectroscopy as a promising miniaturized and high throughput portable spectral imaging technique suffers from the problem of low precision and slow speed, which limits its wide use in various applications. To reconstruct the diffractive spectral image more accurately and fast, a three-dimensional spectrum recovery algorithm is proposed in this paper. The algorithm takes advantage of a neural network for image reconstruction which consists of a U-Net architecture with 3D convolutional layers to improve the processing precision and speed. Numerical experiments are conducted to prove its effectiveness. It is shown that the mean peak signal-to-noise ratio (MPSNR) of the recovered image relative to the original image is improved by 1.8 dB in comparison to other traditional methods. In addition, the obtained mean structural similarity (MSSIM) of 0.91 meets the standard of discrimination to human eyes. Moreover, the algorithm runs in just 0.36 s, which is faster than other traditional methods. 3D convolutional networks play a critical role in performance improvement. Improvements in processing speed and accuracy have greatly benefited the realization and application of diffractive optical imaging spectroscopy. The new algorithm with high accuracy and fast speed has a great potential application in diffraction lens spectroscopy and paves a new way for emerging more portable spectral imaging technique.

Single-pixel imaging (SPI) leverages sequential pattern illumination and intensity detection to reconstruct images, facing the challenge of balancing high-resolution output with ultra-low sampling rates for rapid imaging processes. We introduce a network architecture specifically tailored for SPI, which demonstrates improved performance even before integrating with SPI’s physical sampling processes. This integration, particularly focusing on the nuanced effects of sampling rates within the model’s loss function and data preprocessing, enhances image reconstruction quality and adaptability at low sampling rates, down to 1.56%. Our approach achieves a balance between advanced computational methods and the physical principles of SPI, resulting in a peak signal-to-noise ratio of 30.93 dB, a structural similarity index measure of 0.8818, and a perceptual index (PI) of 5.31 at a 6.25% sampling rate, alongside a notable PI of 2.68 at a 1.56% sampling rate in practical tests. By merging sophisticated network design with strategic integration of physical sampling rates, our model provides a refined solution for high-quality, high-resolution SPI at minimal sampling rates, facilitating progress in ultra-fast imaging applications.

This work aims to simplify the characterization process of coded-apertures for computational imaging (CI) at microwave frequencies. A major benefit of the presented technique is the minimization of the processing time needed to calculate the system sensing matrix for microwave CI-based compressive sensing applications. To achieve this, a deep learning-based approach which is capable of generating the sensing matrix using features learned directly from the coded-aperture distribution is proposed. To avoid the vanishing gradient problem, the proposed deep learning network contains skip connections. Using a dataset of 1,000 testing samples, the average normalized mean-squared-error (NMSE) calculated between the sensing matrix generated by the conventional method and that predicted by the proposed network is 0.0036. Moreover, the average mean-squared-error (MSE) calculated between the images reconstructed using the conventional and the predicted sensing matrix is 0.00297. In addition to providing high-fidelity estimations with minimized error, we demonstrate that using the trained network, the prediction of the sensing matrix can be achieved in 0.212 s, corresponding to a 65% reduction in the computation time needed to calculate the sensing matrix. This has significant outcomes in achieving real-time operation of CI-based microwave imaging systems.

Copyright of Journal of Data Acquisition & Processing / Shu Ju Cai Ji Yu Chu Li is the property of Editorial Department of Journal of Nanjing University of Aeronautics & Astronautics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Data Acquisition & Processing / Shu Ju Cai Ji Yu Chu Li is the property of Editorial Department of Journal of Nanjing University of Aeronautics & Astronautics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Data Acquisition & Processing / Shu Ju Cai Ji Yu Chu Li is the property of Editorial Department of Journal of Nanjing University of Aeronautics & Astronautics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

New computed tomography (CT) algorithms are commonly developed in high-level programming languages, such as Python or MATLAB, while low-level languages are used to support their computation-intensive operations. In the past decade, graphics processing units (GPUs) have become the de-facto standard for large parallel computations in areas such as computational imaging, image processing, and machine learning. Our fast-and-flexible CT reconstruction software, ASTRA Toolbox, therefore already implemented tomographic projectors, i.e., the core computational operations modeling the X-ray physics, using NVIDIA CUDA (Compute Unified Device Architecture), a low-level platform for computation on GPUs. However, the Python-C++ language barrier prevents high-level Python users from modifying these low-level projectors, and, as a consequence, research into new tomographic algorithms is more complex and time-consuming than necessary. With the ASTRA KernelKit, we lifted tomographic projectors to Python and leveraged CuPy, a numerical software like NumPy and SciPy that exposes CUDA to Python, to obtain a fine-grained control over their efficiency and implementation. In this article, we introduced our software and illustrated its importance for high-performance and data-driven applications using examples from deep learning, real-time X-ray CT, and kernel tuning. [ABSTRACT FROM AUTHOR]

Fourier ptychographic microscopy (FPM) emerged as a prominent imaging technique in 2013, attracting significant interest due to its remarkable features such as precise phase retrieval, expansive field of view (FOV), and superior resolution. Over the past decade, FPM has become an essential tool in microscopy, with applications in metrology, scientific research, biomedicine, and inspection. This achievement arises from its ability to effectively address the persistent challenge of achieving a trade-off between FOV and resolution in imaging systems. It has a wide range of applications, including label-free imaging, drug screening, and digital pathology. In this comprehensive review, we present a concise overview of the fundamental principles of FPM and compare it with similar imaging techniques. In addition, we present a study on achieving colorization of restored photographs and enhancing the speed of FPM. Subsequently, we showcase several FPM applications utilizing the previously described technologies, with a specific focus on digital pathology, drug screening, and three-dimensional imaging. We thoroughly examine the benefits and challenges associated with integrating deep learning and FPM. To summarize, we express our own viewpoints on the technological progress of FPM and explore prospective avenues for its future developments. [ABSTRACT FROM AUTHOR]