Search

Emerging dynamic metasurface antennas (DMAs) technology promises to achieve future wireless communications by reducing hardware costs, physical size and power consumption. Especially in the field of radar imaging, they can be used as an effective alternative platform for modern computational imaging; because they can simplify the physical hardware and increase the data acquisition rate. Fourier transform (FT)-based scene image reconstruction techniques are known as cost-effective computing solutions for the imaging system processing unit. However, due to the physical layer compression in DMAs and the fact that they do not produce uniform radiation patterns, the information provided by them is not compatible with Fourier-based techniques and cannot be applied directly. In this paper, we first introduce a 3D near-field bistatic imaging approach using two one-dimensional (1D) DMAs as a panel-to-panel model in a Mills Cross structure. Then, based on the introduced mathematical model, we derive a Fourier-based algorithm for the image reconstruction problem. The proposed algorithm consists of five main steps: pre-processing (to transfer data provided by the transmitter and receiver DMAs to a set of equivalent spatial measurements), applying FTs to the transferred signal, filtering in the Fourier domain, a simplified interpolation, and applying a 3D inverse FT to retrieve scene information. The results of numerical simulations confirm the satisfactory performance of the proposed approach.

The performance of long-range imaging systems often suffers due to the presence of atmospheric turbulence. One way to alleviate the degradation caused by atmospheric turbulence is to apply post-processing mitigation algorithms, where a high-quality frame is reconstructed from a single degraded image or a sequence of degraded frames. The image processing algorithms for atmospheric turbulence mitigation have been studied for decades, yet some critical problems remain open. This dissertation addresses the problem of image reconstruction through atmospheric turbulence from three unique perspectives: 1) Reconstruction with the presence of moving objects using an improved classical image processing pipeline. 2) A fast simulation scheme for efficiently generating large-scale turbulence-degraded datasets for training deep neural networks. 3) A deep learning-based single-frame reconstruction method using Vision Transformer.

Imaging and computer vision systems deployed in real-world environments face the challenge of accommodating a wide range of lighting conditions. However, the cost, the demand for high resolution, and the miniaturization of imaging devices impose physical constraints on sensor design, limiting both the dynamic range and effective aperture size of each pixel. Consequently, conventional CMOS sensors fail to deliver satisfactory capture in high dynamic range scenes or under photon-limited conditions, thereby impacting the performance of downstream vision tasks. In this thesis, we address two key problems: 1) exploring the utilization of spatial multiplexing, specifically spatially varying exposure tiling, to extend sensor dynamic range and optimize scene capture, and 2) developing techniques to enhance the robustness of object detection systems under photon-limited conditions. In addition to challenges imposed by natural environments, real-world vision systems are susceptible to adversarial attacks in the form of artificially added digital content. Therefore, this thesis presents a comprehensive pipeline for constructing a robust and scalable system to counter such attacks.

Coordinate networks are neural networks that parameterize low-dimensional signals like images and signed distance functions. They fail to sequentially learn on a stream of signals as they catastrophically forget the earlier part of the stream. This paper proposes a method to mitigate forgetting based on a hash encoding coordinate network. Specifically, the hash table is augmented with a cache data structure along with additional regularizations. Experiments show that this method eliminates forgetting when the network learns dense signals for different data collection patterns, approaching the empirical upper bound with little margin.

In this paper, we present a hybrid refractive-diffractive lens that, when paired with a deep neural network-based image reconstruction, produces high-quality, real-world images with minimal artifacts, reaching a PSNR of 28 dB on the test set. Our diffractive element compensates for the off-axis aberrations of a single refractive element and has reduced chromatic aberrations across the visible light spectrum. We also describe our training set augmentation and novel quality criteria called “false edge level” (FEL), which validates that the neural network produces visually appealing images without artifacts under a wide range of ISO and exposure settings. Our quality criteria (FEL) enabled us to include real scene images without a corresponding ground truth in the training process.

In this paper, a preliminary study related to the detection of breast cancer based on a computational millimetre-wave (mmW) imaging system is presented at K-band frequencies. In comparison to normal tissues, the different dielectric properties (permittivity and conductivity) of the malignant tissues can be exploited in order to detect the presence of tumour through a mmW imaging system. This work demonstrates the detection of breast tumour as an application of a computational imaging technique by leveraging the concept of the dynamic metasurface antenna (DMA) aperture as a transmitter. In this framework, the computational imaging aperture can provide a compact system design with a low-cost deployment for detecting tumours in early-stage, and this can speed up the screening process of the population at risk. Particularly, early detection of Stage 1 can help with the determination of the best treatment method and enhance a patient’s prognosis.

Environmental scene reconstruction is of great interest for autonomous robotic applications, since an accurate representation of the environment is necessary to ensure safe interaction with robots. Equally important, it is also vital to ensure reliable communication between the robot and its controller. Large Intelligent Surface (LIS) is a technology that has been extensively studied due to its communication capabilities. Moreover, due to the number of antenna elements, these surfaces arise as a powerful solution to radio sensing. This paper presents a novel method to translate radio environmental maps obtained at the LIS to floor plans of the indoor environment built of scatterers spread along its area. The usage of a Least Squares (LS) based method, U-Net (UN) and conditional Generative Adversarial Networks (cGANs) were leveraged to perform this task. We show that the floor plan can be correctly reconstructed using both local and global measurements.

Five established clearing protocols were compared with a modified and simplified method to determine an optimal clearing reagent for three-dimensionally visualizing fluorophores in the murine liver, a challenging organ to clear. We report successful clearing of whole liver lobes by modification of an established protocol (UbasM) using only Ub-1, a urea-based amino sugar reagent, in a simpler protocol that requires only a 24-h processing time. With Ub-1 alone, we observed sufficiently preserved liver tissue structure in three dimensions along with excellent preservation of fluorophore emissions from endogenous protein reporters and lipophilic tracer dyes. This streamlined technique can be used for 3D cell lineage tracing and fluoroprobe-based reporter gene expression to compare various experimental conditions., METHOD SUMMARY This study presents a simplified protocol for optically clearing murine liver tissue in only 24 h using one simple urea-based amino sugar solution and a single incubation. This method preserves fluorescence of transgenically expressed proteins and lipophilic tracer dyes within the context of native spatial morphology., Graphical abstract

Comment: 8 pages, 5 figures, Recovering a signal from auto-correlations or, equivalently, retrieving the phase linked to a given Fourier modulus, is a wide-spread problem in imaging. This problem has been tackled in a number of experimental situations, from optical microscopy to adaptive astronomy, making use of assumptions based on constraints and prior information about the recovered object. In a similar fashion, deconvolution is another common problem in imaging, in particular within the optical community, allowing high-resolution reconstruction of blurred images. Here we address the mixed problem of performing the auto-correlation inversion while, at the same time, deconvolving its current estimation. To this end, we propose an I-divergence optimization, driving our formalism into a widely used iterative scheme, inspired by Bayesian-based approaches. We demonstrate the method recovering the signal from blurred auto-correlations, further analysing the cases of blurred objects and band-limited Fourier measurements.

We study the coherence in time and space of electromagnetic fields propagated through complex media. Whether for localization, imaging or telecommunication, the development of dedicated numerical techniques is generally based on the exploitation of simplified models considering either coherent or diffuse fields. The optimization of such applications in conditions of partial coherence can therefore be particularly challenging, requiring the development of hybrid algorithms adaptable to prior knowledge on the processed fields. The objective of this work is to provide numerical techniques for decomposing an electromagnetic field into subspaces that can then be filtered according to their level of spatial and temporal coherence. In contrast to the studies carried out on space-space transfer matrices notably used for the calculation of Wigner-Smith operators, these decompositions are carried out on space-time matrices in order to facilitate the study of temporal dispersion. The theory is developed for illustrative purposes using experimental results from a leaky resonant system but seem to be applicable to any scattering and reverberating media capable of transforming localized and coherent excitations into complex and diffuse distributions. To conclude this work, the proposed technique is exploited to improve image reconstruction in a millimeter-wave computational imaging demonstration. In the studied context and from a more general perspective, we propose a technique to select the most suitable subspaces for each application operating under conditions of partial coherence, whether these correspond in the most extreme cases to ballistic paths or to diffuse fields., 12 pages, 14 figures

International audience; The development of new imaging systems based on computational techniques greatly alleviate the hardware constraints usually encountered in these domains. The limitations associated with the use of highly redundant active chains are thus substituted by the development of radiating systems allowing the transfer of these limitations to the software layer. Faced with the large volumes of data to be processed, it is still necessary to propose numerical techniques to limit the size of the problems to be solved while ensuring that the imaging results remain exploitable for the targeted applications. Principal component analysis offers interesting insights in such a context.

Signal retrieval from a series of indirect measurements is a common task in many imaging, metrology and characterization platforms in science and engineering. Because most of the indirect measurement processes are well-described by physical models, signal retrieval can be solved with an iterative optimization that enforces measurement consistency and prior knowledge on the signal. These iterative processes are time-consuming and only accommodate a linear measurement process and convex signal constraints. Recently, neural networks have been widely adopted to supersede iterative signal retrieval methods by approximating the inverse mapping of the measurement model. However, networks with deterministic processes have failed to distinguish signal ambiguities in an ill-posed measurement system, and retrieved signals often lack consistency with the measurement. In this work we introduce a variational generative model to capture the distribution of all possible signals, given a particular measurement. By exploiting the known measurement model in the variational generative framework, our signal retrieval process resolves the ambiguity in the forward process, and learns to retrieve signals that satisfy the measurement with high fidelity in a variety of linear and nonlinear ill-posed systems, including ultrafast pulse retrieval, coded aperture compressive video sensing and image retrieval from Fresnel hologram., 8 pages, 5 figures. Initial submission to IEEE Transactions on Computational Imaging

An ultra-high-speed computational CMOS image sensor with a burst frame rate of 303 megaframes per second, which is the fastest among the solid-state image sensors, to our knowledge, is demonstrated. This image sensor is compatible with ordinary single-aperture lenses and can operate in dual modes, such as single-event filming mode or multi-exposure imaging mode, by reconfiguring the number of exposure cycles. To realize this frame rate, the charge modulator drivers were adequately designed to suppress the peak driving current taking advantage of the operational constraint of the multi-tap charge modulator. The pixel array is composed of macropixels with 2 × 2 4-tap subpixels. Because temporal compressive sensing is performed in the charge domain without any analog circuit, ultrafast frame rates, small pixel size, low noise, and low power consumption are achieved. In the experiments, single-event imaging of plasma emission in laser processing and multi-exposure transient imaging of light reflections to extend the depth range and to decompose multiple reflections for time-of-flight (TOF) depth imaging with a compression ratio of 8× were demonstrated. Time-resolved images similar to those obtained by the direct-type TOF were reproduced in a single shot, while the charge modulator for the indirect TOF was utilized.

Rapid advances in neuroimaging have resulted in large repositories of images with high temporal and spatial resolutions. This has motivated large-scale connectomic analyses aimed at understanding representation and processing of stimuli. Such analyses require novel computational tools that significantly extend the state-of-the-art in machine learning and data science. These studies have far-reaching implications for a variety of scientific fields, including but not limited to (i) neuroscience, the study of computational, theoretical, behavioural and cellular aspects of the functioning of brains(ii) neuropathology, the branch of medicine dedicated to the study of brain diseases and disorders(iii) psychiatry, the branch of medicine focused on diagnosis, treatment and prevention of mental, emotional and behavioural disorders, and (iv) the design of human-computer interfaces, specifically brain-computer interfaces in advanced augmented reality (AR), virtual reality (VR), and assistive technologies such as robotic exoskeletons and prosthetics. In this thesis, we will focus our analyses on functional neuroimages (fMRIs). We will discuss three related, yet distinct threads of work. First, we will present the problem of identifiability, where we present our method to show that two distinct fMRIs taken over different sessions of the same subject are more similar in terms of their connectomes than those from two different subjects. We will present new results that identify specific parts of resting state and task-specific connectomes that are responsible for the unique \emph{signatures}. We show that a very small part of the connectome can be used to derive features for discriminating between individuals. A network of these features is shown to achieve excellent training and test accuracy in matching imaging datasets. We show that these features are statistically significant, robust to perturbations, invariant across populations, and are localized to a small number of structural regions of the brain. Furthermore, we show that for task-specific connectomes, the regions identified by our method are consistent with their known functional characterization. We present a matrix sampling technique to derive computationally efficient and accurate methods for identifying the discriminating sub-connectome and support all of the claims using state-of-the-art statistical tests and computational techniques. We will discuss the privacy implications of brain signatures. We present a de-anonymization attack rooted in the innate uniqueness of the structure and function of the human brain. We show that the attack reveals not only the {\em identity of an individual, but also the efficacy with which they performing cognitive tasks}. Our attack relies on a matrix analyses techniques that are used to extract discriminating features in neuroimages. We will also discuss challenges associated with defending against such attacks. The second thread of projects deal with network analyses of functional brain images. We model functional connectomes as networks where nodes have spatial coordinates associated with them. The problem of identifying brain signatures using such networks maps to the problem of identifying conserved patterns corresponds to the alignment of the networks. While traditional methods tackle the problem of alignment using structure obtained from images, we formulate a novel problem -- {\em rigid graph alignment}, which simultaneously aligns the network, as well as the underlying structure. We formally specify the problem and present a method based on alternating least squares which alternately aligns the network and the structure via rigid body transformations. We demonstrate that this approach significantly improves the quality of network alignment in synthetic graphs as well as in brain networks. The power of this approach is that it can be applied to any graph where nodes have spatial coordinates, and that it works very well with a number of state-of-the-art network aligners that admit a prior. Next, we model brain activity as dynamic correlation graphs. An important problem in the context of such dynamic correlation graphs is the discovery of sets of regions of the brain, whose activity level is temporally coherent. These manifest as temporally persistent sub-graphs that are strongly connected, referred to as coherent subgraphs. We present a model and method for identifying coherent subgraphs in dynamic correlation graphs. We show that densely connected components in correlation graphs can be effectively modeled as low-rank sub-matrices derived from the time series signals. Specifically, we derive theoretical results showing that quasi-cliques in a correlation graph can be inferred from rows of the left singular matrix of its time-series, and can be tracked in time to identify coherent subgraphs. We apply the proposed method to real-world time-series data from functional MRIs to show that signals corresponding to nodes in coherent subgraphs can accurately predict whether the subject was actively performing a cognitive task, or was at rest. We also show that the same set of nodes can predict task outcomes/ conditions (such as win v/s loss in a gambling task). In the final thread of projects, we develop methods to understand how humans process naturalistic visual inputs. A major problem associated with the analyses of such as audiovisual input from videos (eg, movies), is that observed neuronal responses are due to combinations of “pure” factors, many of which may be latent. We present a novel methodological framework for deconvolving the brain’s response to mixed stimuli into its constituent responses to underlying pure factors. This framework, based on archetypal analysis, is applied to the analysis of imaging data from an adult cohort watching the BBC show, Sherlock. By focusing on visual stimulus, we show strong correlation between the observed deconvolved response and third party textual video annotations–demonstrating the significant power of the analyses techniques. Building on these results, we show that this techniques can be used to predict neuronal responses in new subjects (how other individuals react to Sherlock), as well as to new visual content (how individuals react to other videos with known annotations). Next, we describe a novel effort aimed at reconstructing video frames from observed functional MRI (fMRI) images. We demonstrate that our model can predict visual objects (such as face, car, etc.) and abstract aspects relating to scenes (such as indoor/outdoor). Our multi-modal learning approach first learns a representation of the video frames using an encoder-decoder model, and then learns a mapping from observed fMRI response to corresponding latent video frame representation. We demonstrate the power of our model and methods using a number of tests: (i) we show that the latent representations of video frames and those constructed from corresponding fMRI images are highly clustered; (ii) the latent representations can be used to predict objects in video frames using just the fMRI frames with good accuracy; and (iii) fMRI responses can be used to reconstruct the original inputs with high fidelity to predict the presence or absence of objects from corresponding frames.

A refractive lens is one of the simplest, most cost-effective and easily available imaging elements. Given a spatially incoherent illumination, a refractive lens can faithfully map every object point to an image point in the sensor plane, when the object and image distances satisfy the imaging conditions. However, static imaging is limited to the depth of focus, beyond which the point-to-point mapping can only be obtained by changing either the location of the lens, object or the imaging sensor. In this study, the depth of focus of a refractive lens in static mode has been expanded using a recently developed computational reconstruction method, Lucy-Richardson-Rosen algorithm (LRRA). The imaging process consists of three steps. In the first step, point spread functions (PSFs) were recorded along different depths and stored in the computer as PSF library. In the next step, the object intensity distribution was recorded. The LRRA was then applied to deconvolve the object information from the recorded intensity distributions during the final step. The results of LRRA were compared with two well-known reconstruction methods, namely the Lucy-Richardson algorithm and non-linear reconstruction.

This article presents a holographic metasurface antenna with stochastically distributed surface impedance, which produces randomly frequency-diverse radiation patterns. Low mutual coherence electric field patterns generated by the holographic metasurface antenna can cover the K-band from 18 to 26 GHz with 0.1 GHz intervals. By utilizing the frequency-diverse holographic metasurface (FDHM) antenna, we build a near-field microwave computational imaging system based on reflected signals in the frequency domain. A standard horn antenna is adopted to acquire frequency domain signals radiated from the proposed FDHM antenna. A detail imaging restoration process is presented, and the desired targets are correctly reconstructed using the 81 frequency-diverse patterns through full-wave simulation studies. Compressed sensing technique and iterative shrinkage/thresholding algorithms are applied for the imaging reconstruction. The achieved compressive ratio of this computational imaging system on the physical layer is 30:1.

Significance: Fourier ptychography microscopy (FPM) is a computational optical imaging technology that employs angularly varying illuminations and a phase retrieval algorithm to achieve a wide field of view and high-resolution imaging simultaneously. In the FPM, LED position error will reduce the quality of the reconstructed high-resolution image. To correct the LED positions, current methods consider each of the LED positions as independent and use an optimization algorithm to get each of the positions. When the positional misalignment is large or the search position falls into a local optimal value, the current methods may lack stability and accuracy. Aim: We improve the model of the LED position and propose an accurate and stable two-step correction scheme (tcFPM) to calibrate the LED position error. Approach: The improved LED positions model combines the overall offset, which represents the relative deviation of the LED array and the optical axis, with the slight deviation of each LED’s independent position. In the tcFPM, the overall offset of the LED array is corrected at first, which obtains an approximate value of the overall offset of the LED array. Then the position of each LED is precisely adjusted, which obtains the slight offset of each LED. Results: This LED position error model is more in line with the actual situation. The simulation and experimental results show that the method has high accuracy in correcting the LED position. Furthermore, the reconstruction process of tcFPM is more stable and significantly improves the quality of the reconstruction results, which is compared with some LED position error correction methods. Conclusions: An LED position error correction technology is proposed, which has a stable iterative process and improves the reconstruction accuracy of complex amplitude.

Direct imaging systems that create an image of an object directly on the sensor in a single step are prone to many constraints, as a perfect image is required to be recorded within this step. In designing high resolution direct imaging systems with a diffractive lens, the outermost zone width either reaches the lithography limit or the diffraction limit itself, imposing challenges in fabrication. However, if the imaging mode is switched to an indirect one consisting of multiple steps to complete imaging, then different possibilities open. One such method is the widely used indirect imaging method with Golay configuration telescopes. In this study, a Golay-like configuration has been adapted to realize a large-area diffractive lens with three sub-aperture diffractive lenses. The sub-aperture diffractive lenses are not required to collect light and focus them to a single point as in a direct imaging system, but to focus independently on different points within the sensor area. This approach of a Large-Area Diffractive lens with Integrated Sub-Apertures (LADISA) relaxes the fabrication constraints and allows the sub-aperture diffractive elements to have a larger outermost zone width and a smaller area. The diffractive sub-apertures were manufactured using photolithography. The fabricated diffractive element was implemented in indirect imaging mode using non-linear reconstruction and the Lucy–Richardson–Rosen algorithm with synthesized point spread functions. The computational optical experiments revealed improved optical and computational imaging resolutions compared to previous studies.

The ability to sculpt complex reference waves and probe diverse radiation field patterns have facilitated the rise of metasurface antennas, while there is still a compromise between the required wide operation band and the non-overlapping characteristic of radiation field patterns. Specifically, the current computational image formation process with a classic matched filter and other sparsity-driven algorithms would inevitably face the challenge of a relatively confined scene information sampling ratio and high computational complexity. In this paper, we marry the concepts of a deep convolutional neural network with computational imaging literature. Compared with the current matched filter and compressed sensing reconstruction technique, our proposal could handle a relatively high correlation of measurement modes and low scene sampling ratio. With the delicately trained reconstruction network, point-size objects and more complicated targets can both be quickly and accurately reconstructed. In addition, the unavoidable heavy computation burden and essential large operation frequency band can be effectively mitigated. The simulated experiments with measured radiation field data verify the effectiveness of the proposed method.

Imaging systems with miniaturized device footprint, real-time processing speed and high resolution three-dimensional (3D) visualization are critical to broad biomedical applications such as endoscopy. Most of existing imaging systems rely on bulky lenses and mechanically refocusing to perform 3D imaging. Here, we demonstrate GEOMScope, a lensless single-shot 3D microscope that forms image through a single layer of thin microlens array and reconstructs objects through an innovative algorithm combining geometrical-optics-based pixel back projection and background suppressions. We verify the effectiveness of GEOMScope on resolution target, fluorescent particles and volumetric objects. Comparing to other widefield lensless imaging devices, we significantly reduce the required computational resource and increase the reconstruction speed by orders of magnitude. This enables us to image and recover large volume 3D object in high resolution with near real-time processing speed. Such a low computational complexity is attributed to the joint design of imaging optics and reconstruction algorithms, and a joint application of geometrical optics and machine learning in the 3D reconstruction. More broadly, the excellent performance of GEOMScope in imaging resolution, volume, and reconstruction speed implicates that geometrical optics could greatly benefit and play an important role in computational imaging.

The image-based discourse on clay figurines that treated them as merely artistic representations, the meaning of which needs to be deciphered through various iconological methods, has been severely critiqued and challenged in the past decade. This discourse, however, has largely shaped the way that figurines are depicted in archaeological iterations and publications, and it is this corpus of images that has in turn shaped further thinking and discussion on figurines, especially since very few people are able to handle the original, three-dimensional, physical objects. Building on the changing intellectual climate in figurine studies, we propose here a framework that treats figurines as multi-sensorial, affective and dynamic objects, acting within distinctive, relational fields of sensoriality. Furthermore, we situate a range of digital, computational methods within this framework in an attempt to deprive them of their latent Cartesianism and mentalism, and we demonstrate how we have applied them to the study of Neolithic figurines from the site of Koutroulou Magoula in Greece. We argue that such methodologies, situated within an experiential framework, not only provide new means of understanding, interpretation and dissemination, but, most importantly, enable researchers and the public to explore the sensorial affordances and affective potential of things, in the past as well as in the present.

A retina-like 3-D (three-dimensional) computational ghost imaging (R-3DCGI) is proposed to obtain high resolution while maintaining high imaging efficiency. R-3DCGI models are developed, and corresponding simulations are conducted to verify and evaluate the accuracy and features of the R-3DCGI. Results show that the models obtain rotation and scaling invariances and compress redundant information, thereby improving imaging efficiency over the traditional computational ghost imaging. The relationships between parameters of the retina-like models and signal-noise ratio are discussed to optimize the retina-like structures in various applications.

A new method of creating microwave imagery by capturing the signals emitted by a small set of wireless WiFi transmitters is presented. The imaging technique leverages the fact that the signals emitted by separate WiFi transmitters are sufficiently statistically independent to create a radiation pattern that is spatially incoherent, enabling the use of spatial frequency sampling using a small set of receiving antennas in a sparse array. In contrast to traditional microwave imaging, this method requires no mechanical or electrical beam scanning and no coordination between transmitters and receivers. Furthermore, the WiFi imaging system requires far less receiver gain than passive microwave imagers and significantly less bandwidth. We experimentally demonstrate the 2-D image reconstruction of reflecting metal spheres and an X-shaped reflecting target using three transmitters emitting independent 16-level quadrature amplitude modulated signals at 5.5 GHz matching commercial WiFi protocols 802.11n/ac.

3D Fourier ptychographic microscopy can recover objects with high resolution across large volumes, which is very challenging in 3D imaging particularly because multiple scattering may occur within the sample. In this paper, we assume the thick sample is composed of a number of thin slices, and employ the beam propagation method (BPM) to model the propagation process of the light wave among successive slices capturing multiple scattering effects. 3D imaging is then accomplished by utilizing the gradient descent method to minimize the difference between the estimated intensity images by BPM and the captured measurements with angle-varied illuminations. We adopt a time-reversal scheme to obtain the gradient of the transmitted light intensity with respect to the complex refractive index of the volumetric sample and use the error backpropagation method to update the 3D sample iteratively. To further preserve the sharpness of the edges, we introduce a sparsity regularization term into the optimization process. Our method can achieve higher performance of 3D reconstruction compared to the original multi-slice approach, demonstrated in both simulation and experiment.

We investigated optimal implementation strategies for industrial inspection systems aiming to detect cracks on ground steel billets’ surfaces by combining state-of-the-art AI-based methods and classical computational imaging techniques. In 2D texture images, the interesting patterns of surface irregularities are often surrounded by visual clutter, which is to be ignored, e.g., grinding patterns. Even neural networks struggle to reliably distinguish between actual surface disruptions and irrelevant background patterns. Consequently, the image acquisition procedure already has to be optimised to the specific application. In our case, we use photometric stereo (PS) imaging to generate 3D surface models of steel billets using multiple illumination units. However, we demonstrate that the neural networks, especially in high-speed scenarios, still suffer from recognition deficiencies when using raw photometric stereo camera data, and are unable to generalise to new billets and image acquisition conditions. Only the additional application of adequate state-of-the-art image processing algorithms guarantees the best results in both aspects. The neural networks benefit when appropriate image acquisition methods together with image processing algorithms emphasise relevant surface structures and reduce overall pattern variation. Our proposed combined strategy shows a 9.25% better detection rate on validation data and is 14.7% better on test data, displaying the best generalisation.

The special dispersion and temperature characteristics of diffractive optical element (DOE) make them widely used in optical systems that require both athermalization and achromatic aberrations designs. The multi-layer DOE (MLDOE) can improve the diffraction efficiency of the overall broad waveband, but its diffraction efficiency decreases with changes in ambient temperature. When the ambient temperature changes, the micro-structure heights of MLDOE and the refractive index of the substrate materials change, ultimately affecting its diffraction efficiency, and, further, the optical transform function (OTF). In this paper, the influence of ambient temperature on the diffraction efficiency of MLDOE in a dual-infrared waveband is proposed and discussed, the diffraction efficiency of MLDOE caused by ambient temperature is derived, and a computational imaging method that combines optical design and image restoration is proposed. Finally, a dual-infrared waveband infrared optical system with athermalization and achromatic aberrations corrected based on computational imaging method is designed. Results show that this method can effectively reduce the diffraction efficiency of MLDOE by ambient temperature and improve the imaging quality of hybrid optical systems.

The scattering of light after passing through a complex medium poses challenges in many fields. Any point in the collected speckle will contain information from the entire target plane because of the randomness of scattering. The detailed information of complex targets is submerged in the aliased signal caused by random scattering, and the aliased signal causes the quality of the recovered target to be degraded. In this paper, a new neural network named Adaptive Encoding Scattering Imaging ConvNet (AESINet) is constructed by analyzing the physical prior of speckle image redundancy to recover complex targets hidden behind the opaque medium. AESINet reduces the redundancy of speckle through adaptive encoding which effectively improves the separability of data; the encoded speckle makes it easier for the network to extract features, and helps restore the detailed information of the target. The necessity for adaptive encoding is analyzed, and the ability of this method to reconstruct complex targets is tested. The peak signal-to-noise ratio (PSNR) of the reconstructed target after adaptive encoding can be improved by 1.8 dB. This paper provides an effective reference for neural networks combined with other physical priors in scattering processes.

Background: Population-based state cancer registries are an authoritative source for cancer statistics in the United States. They routinely collect a variety of data, including patient demographics, primary tumor site, stage at diagnosis, first course of treatment, and survival, on every cancer case that is reported across all U.S. states and territories. The goal of our project is to enrich NCI’s Surveillance, Epidemiology, and End Results (SEER) registry data with high-quality population-based biospecimen data in the form of digital pathology, machine-learning-based classifications, and quantitative histopathology imaging feature sets (referred to here as Pathomics features). Materials and Methods: As part of the project, the underlying informatics infrastructure was designed, tested, and implemented through close collaboration with several participating SEER registries to ensure consistency with registry processes, computational scalability, and ability to support creation of population cohorts that span multiple sites. Utilizing computational imaging algorithms and methods to both generate indices and search for matches makes it possible to reduce inter- and intra-observer inconsistencies and to improve the objectivity with which large image repositories are interrogated. Results: Our team has created and continues to expand a well-curated repository of high-quality digitized pathology images corresponding to subjects whose data are routinely collected by the collaborating registries. Our team has systematically deployed and tested key, visual analytic methods to facilitate automated creation of population cohorts for epidemiological studies and tools to support visualization of feature clusters and evaluation of whole-slide images. As part of these efforts, we are developing and optimizing advanced search and matching algorithms to facilitate automated, content-based retrieval of digitized specimens based on their underlying image features and staining characteristics. Conclusion: To meet the challenges of this project, we established the analytic pipelines, methods, and workflows to support the expansion and management of a growing repository of high-quality digitized pathology and information-rich, population cohorts containing objective imaging and clinical attributes to facilitate studies that seek to discriminate among different subtypes of disease, stratify patient populations, and perform comparisons of tumor characteristics within and across patient cohorts. We have also successfully developed a suite of tools based on a deep-learning method to perform quantitative characterizations of tumor regions, assess infiltrating lymphocyte distributions, and generate objective nuclear feature measurements. As part of these efforts, our team has implemented reliable methods that enable investigators to systematically search through large repositories to automatically retrieve digitized pathology specimens and correlated clinical data based on their computational signatures.

In this roadmap article, we have focused on the most recent advances in terahertz (THz) imaging with particular attention paid to the optimization and miniaturization of the THz imaging systems. Such systems entail enhanced functionality, reduced power consumption, and increased convenience, thus being geared toward the implementation of THz imaging systems in real operational conditions. The article will touch upon the advanced solid-state-based THz imaging systems, including room temperature THz sensors and arrays, as well as their on-chip integration with diffractive THz optical components. We will cover the current-state of compact room temperature THz emission sources, both optolectronic and electrically driven; particular emphasis is attributed to the beam-forming role in THz imaging, THz holography and spatial filtering, THz nano-imaging, and computational imaging. A number of advanced THz techniques, such as light-field THz imaging, homodyne spectroscopy, and phase sensitive spectrometry, THz modulated continuous wave imaging, room temperature THz frequency combs, and passive THz imaging, as well as the use of artificial intelligence in THz data processing and optics development, will be reviewed. This roadmap presents a structured snapshot of current advances in THz imaging as of 2021 and provides an opinion on contemporary scientific and technological challenges in this field, as well as extrapolations of possible further evolution in THz imaging.

In recent years, lensfree on-chip microscopy has developed into a promising and powerful computational optical microscopy technique that allows for wide-field, high-throughput microscopic imaging without using any lenses. However, due to the limited pixel size of the state-of-the-art image sensors, lens-free on-chip microscopy generally suffers from low imaging resolution, which is far from enough to meet the current demand for high-resolution microscopy. Many pixel super-resolution techniques have been developed to solve or at least partially solve this problem by acquiring a series of low-resolution holograms with multiple lateral sub-pixel shifting or axial distances. However, the prerequisite of these pixel super-resolution techniques is that the propagation distance of each low-resolution hologram can be obtained precisely, which faces two major challenges. On the one hand, the captured hologram is inherent pixelated and of low resolution, making it difficult to determine the focal plane by evaluating the image sharpness accurately. On the other hand, the twin-image is superimposed on the backpropagated raw hologram, further exacerbating the difficulties in accurate focal plane determination. In this study, we proposed a high-precision autofocusing algorithm for multi-height pixel-super-resolved lensfree on-chip microscopy. Our approach consists of two major steps: individual preliminary estimation and global precise estimation. First, an improved critical function that combines differential critical function and frequency domain critical function is proposed to obtain the preliminary focus distances of different holograms. Then, the precise focus distances can be determined by further evaluating the global offset of the averaged, low-noise reconstruction from all backpropagated holograms with preliminary focus distances. Simulations and experimental results verified the validity and effectiveness of the proposed algorithm.

Significance: Large space-bandwidth product is highly desirable in many biomedical imaging. Fourier ptychographic microscopy (FPM) is a computational imaging technique that can significantly increase the space-bandwidth product of a standard microscope. The illuminator of a Fourier ptychographic microscope is not flexible at present, and it is inconvenient to meet different imaging needs. Aim: An illuminator based on a two-axis motorized rotation stage was presented to provide a more flexible illuminating way with the goal of meeting different imaging needs. Approach: The illuminator adopts a concentric illuminating method to provide coherent illumination in any direction on the sample plane. The sampling pattern can be freely designed and changed according to the parameters of the imaging system. A dither removing algorithm was proposed to remove the potential dither influence introduced in the image acquisition process. Results: The illuminator could be conveniently integrated into different imaging systems. The feasibility and flexibility were demonstrated by applying it to imaging systems with numerical aperture of 0.045 and 0.01. The resolution gain is about 4- and 13-fold, respectively. The effectiveness of the dither removing algorithm was validated in both simulation and experiment. Conclusions: A more flexible illuminator for FPM was presented to meet different imaging needs. A dither removing algorithm was proposed to remove dither influence.

Significance Non–line-of-sight (NLOS) imaging can recover details of a hidden scene from the indirect light that has scattered multiple times. Despite recent advances, NLOS imaging has remained at short-range verifications. Here, both experimental and conceptual innovations yield hardware and software solutions to increase NLOS imaging from meter to kilometer range. This range is about three orders of magnitude longer than previous experiments. The results will open avenues for the development of NLOS imaging techniques and relevant applications to real-world conditions., Non–line-of-sight (NLOS) imaging has the ability to reconstruct hidden objects from indirect light paths that scatter multiple times in the surrounding environment, which is of considerable interest in a wide range of applications. Whereas conventional imaging involves direct line-of-sight light transport to recover the visible objects, NLOS imaging aims to reconstruct the hidden objects from the indirect light paths that scatter multiple times, typically using the information encoded in the time-of-flight of scattered photons. Despite recent advances, NLOS imaging has remained at short-range realizations, limited by the heavy loss and the spatial mixing due to the multiple diffuse reflections. Here, both experimental and conceptual innovations yield hardware and software solutions to increase the standoff distance of NLOS imaging from meter to kilometer range, which is about three orders of magnitude longer than previous experiments. In hardware, we develop a high-efficiency, low-noise NLOS imaging system at near-infrared wavelength based on a dual-telescope confocal optical design. In software, we adopt a convex optimizer, equipped with a tailored spatial–temporal kernel expressed using three-dimensional matrix, to mitigate the effect of the spatial–temporal broadening over long standoffs. Together, these enable our demonstration of NLOS imaging and real-time tracking of hidden objects over a distance of 1.43 km. The results will open venues for the development of NLOS imaging techniques and relevant applications to real-world conditions.

Compressed sensing (CS) is a signal processing approach that solves ill-posed inverse problems, from under-sampled data with respect to the Nyquist criterium. CS exploits sparsity constraints based on the knowledge of prior information, relative to the structure of the object in the spatial or other domains. It is commonly used in image and video compression as well as in scientific and medical applications, including computed tomography and magnetic resonance imaging. In the field of fluorescence microscopy, it has been demonstrated to be valuable for fast and high-resolution imaging, from single-molecule localization, super-resolution to light-sheet microscopy. Furthermore, CS has found remarkable applications in the field of mesoscopic imaging, facilitating the study of small animals' organs and entire organisms. This review article illustrates the working principles of CS, its implementations in optical imaging and discusses several relevant uses of CS in the field of fluorescence imaging from super-resolution microscopy to mesoscopy.

Helical point spread functions (PSFs) provide a powerful computational imaging tool for modern optical imaging and sensing applications. However, their utilization is, so far, limited to a single field of application, i.e. super-resolution microscopy, which is due to multiple shortcomings in their current system implementation. A new computational imaging approach is developed in this thesis, which enables the utilization of helical PSFs and their unique advantages for applications in the area of machine vision. In particular, the approach can be used to acquire the three-dimensional distribution of a passively illuminated, extended scene in a single shot based on a compact, monocular camera setup. A novel image processing routine is established to overcome a major challenge of computational imaging using helical PSFs, i.e. the retrieval of the PSF rotation angle in the case of an extended object distribution. The hardware implementation of computational imaging setups that rely on helical PSFs is based on a combination of a conventional optical element, such as a microscope objective or a camera lens, and an additional, dedicated pupil mask. This mask is commonly realized using either a spatial light modulator or a lithographic element that features a structured surface profile. Two new fabrication schemes with different advantages are explored in this thesis. The first scheme utilizes wafer-scale optical lithography in combination with UV-replication in order to fabricate highly cost efficient phase elements. The second method is based on femto-second laser direct writing. It enables the inscription of the phase element directly inside a transparent optical element using a single fabrication step. Therefore, it facilitates a flexible realization of highly integrated PSF engineered optical systems. Current design concepts for pupil masks that generate helical PSFs only focus on doublehelix distributions that feature two, laterally separated irradiance peaks. Furthermore, a diffraction limited performance of the computational imaging system is assumed. A new design method that enables the generation of multi-order-helix PSFs with an arbitrary number of rotating peaks is developed in this thesis. A study of the influence of first order aberrations on the rotation angle of multi-order-helix PSFs is performed in order to assess their effect on the accuracy limits with respect to three-dimensional imaging. In this context, the superior aberration robustness of high-order-helix PSFs featuring three or more rotating spots is demonstrated. Whereas, on the one hand, the effect of aberrations on helical PSFs degrade the depth retrieval accuracy of three-dimensional imaging systems, their influence can be explored in order to obtain information on the system’s wavefront aberrations on the other hand. To this end, the computational imaging approach developed for three-dimensional imaging is extended and combined with a conventional phase diversity method. The novel approach enables a numerically efficient estimation of general wavefront aberrations based on the acquisition of an extended, unknown object scene. In summary, the research performed in this thesis provides the foundation to exploit the unique advantages of computational imaging systems based on helical PSFs for applications in the area of three-dimensional imaging and wavefront sensing.

Diffractive optical elements (DOEs) are difficult to apply in natural scenes imaging covering the visible bandwidth-spectral due to their strong chromatic aberration and the decrease in diffraction efficiency. Advances in computational imaging make it possible. In this paper, the image quality degradation model of DOE in bandwidth-spectral imaging is established to quantitatively analyze its degradation process. We design a DDZMR network for a single-lens diffractive lens computational imaging system, which can simultaneously perform image restoration and image super-resolution reconstruction on degraded images. The multimodal loss function was created to evaluate the reconstruction of the diffraction imaging degradation by the DDZMR network. The prototype physical prototype of the single-lens harmonic diffraction computational imaging system (SHDCIS) was built to verify the imaging performance. SHDCIS testing showed that optical chromatic aberration is corrected by computational reconstruction, and the computational imaging module can interpret an image and restore it at 1.4 times the resolution. We also evaluated the performance of the DDZMR model using the B100 and Urban100 datasets. Mean Peak Signal to Noise Ratio (PSNR)/Structural Similarity (SSIM) were, respectively, 32.09/0.8975 and 31.82/0.9247, which indicates that DDZMR performed comparably to the state-of-the-art (SOTA) methods. This work can promote the development and application of diffractive imaging systems in the imaging of natural scenes in the bandwidth-spectrum.

The recent multiplication of open access initiatives to Big Data from Space is giving momentum to the field by widening substantially the spectrum of scientific communities and users, as well as awareness among the public, while offering new benefits at all levels from individual citizens to the whole society. Following a detailed and rigorous review process, 14 articles have been selected out of 48 submissions for this special issue. These are briefly summarized.

This thesis investigates the reduction of depth-of- field for miniature camera systems, such as the ones embedded in smartphones. The work on digital shallow depth-of- field is already mature and implemented in most modern phones in the form of image processing algorithms. Yet, little research has been performed to provide optical design solutions for depth-of- field management. This thesis proposes a theoretical framework for the understanding of depth-of-fi eld in the particular case of miniature cameras and two practical solutions to reduce the depth of eld. The fi rst one is an apodization phase-mask designed in order to achieve depth super-resolution with about 20% lobe width reduction. The second one is an f/1.0 and an f/1.4 lens. Both solutions are presented and compared. Finally, some preliminary work is presented on how image processing could further improve the hardware contribution of this thesis.

We examine the nature of the reconstructed 3D image as obtained by replay (or back-propagation) of the object wave from the hologram recording plane to the original object volume. While recording of a hologram involves transferring information from a 3D volume to a 2D detector, the replay of the hologram involves creating information in a set of 3D voxels from a much smaller number of 2D detector pixels, which on a first look appears to be surprising. We point out that the hologram replay process is a Hermitian transpose (and not inverse) of the hologram formation process and therefore only provides an approximation to the original 3D object function. With the knowledge of this Hermitian transpose property, we show how one may realize true 3D image reconstruction via a regularized optimization algorithm. The numerical illustrations of this optimization approach as presented here show excellent slice-by-slice tomographic 3D reconstruction of the original object under the weak scattering approximation. In particular, the reconstructed 3D image field has near-zero numerical values at voxels where the original object did not exist. We note that 3D image reconstruction of this kind cannot be achieved by the traditional physical replay process. In this sense, the proposed methodology for digital holographic image reconstruction goes beyond numerically mimicking the physical process involved in traditional film based holographic replay. The reconstruction approach may find potential applications in a number of digital holographic imaging systems.

Within the framework of Internet of Things or when constrained in limited space, lensless imaging technology provides effective imaging solutions with low cost and reduced size prototypes. In this paper, we proposed a method combining deep learning with lensless coded mask imaging technology. After replacing lenses with the coded mask and using the inverse matrix optimization method to reconstruct the original scene images, we applied FCN-8s, U-Net, and our modified version of U-Net, which is called Dense-U-Net, for post-processing of reconstructed images. The proposed approach showed supreme performance compared to the classical method, where a deep convolutional network leads to critical improvements of the quality of reconstruction.

Label-free, visible light microscopy is an indispensable tool for studying biological nanoparticles (BNPs). However, conventional imaging techniques have two major challenges: (i) weak contrast due to low-refractive-index difference with the surrounding medium and exceptionally small size and (ii) limited spatial resolution. Advances in interferometric microscopy have overcome the weak contrast limitation and enabled direct detection of BNPs, yet lateral resolution remains as a challenge in studying BNP morphology. Here, we introduce a wide-field interferometric microscopy technique augmented by computational imaging to demonstrate a 2-fold lateral resolution improvement over a large field-of-view (>100 × 100 μm2 ), enabling simultaneous imaging of more than 104 BNPs at a resolution of ∼150 nm without any labels or sample preparation. We present a rigorous vectorial-optics-based forward model establishing the relationship between the intensity images captured under partially coherent asymmetric illumination and the complex permittivity distribution of nanoparticles. We demonstrate high-throughput morphological visualization of a diverse population of Ebola virus-like particles and a structurally distinct Ebola vaccine candidate. Our approach offers a low-cost and robust label-free imaging platform for high-throughput and high-resolution characterization of a broad size range of BNPs.

We introduce single-shot X-ray tomography that aims to estimate the target image from a single cone-beam projection measurement. This linear inverse problem is extremely under-determined since the measurements are far fewer than the number of unknowns. Moreover, it is more challenging than conventional tomography where a sufficiently large number of projection angles forms the measurements, allowing for a simple inversion process. However, single-shot tomography becomes less severe if the target image is only composed of known shapes. Hence, the shape prior transforms a linear ill-posed image estimation problem to a non-linear problem of estimating the roto-translations of the shapes. In this paper, we circumvent the non-linearity by using a dictionary of possible roto-translations of the shapes. We propose a convex program CoShaRP to recover the dictionary-coefficients successfully. CoShaRP relies on simplex-type constraint and can be solved quickly using a primal-dual algorithm. The numerical experiments show that CoShaRP recovers shapes stably from moderately noisy measurements., Comment: Paper is currently under consideration for Pattern Recognition Letters

In this work we propose Jacobi–Fourier phase masks for wavefront coding-based imaging systems. The optical properties of the phase mask is study in detail and numerical simulation are shown. Pixel size and noise are taken into account for the deconvolution of images. Numerical simulations indicate that overall performance is better than of the well-known and commonly used trefoil phase This work was supported by the Spanish Ministry of Economía y Competitividad FIS2016-77319-C2-1-R, and FEDER, Xunta de Galicia/FEDER ED431E 2018/08. E. González Amador thanks to Consejo Nacional de Ciencia y Tecnología (CONACyT); with CVU no. 714742. Also, we thank by the support to PADES program; Award no. 2018-13-011-047 SI

By co-designing a meta-optical front end in conjunction with an image-processing back end, we demonstrate noise sensitivity and compactness substantially superior to either an optics-only or a computation-only approach, illustrated by two examples: subwavelength imaging and reconstruction of the full polarization coherence matrices of multiple light sources. Our end-to-end inverse designs couple the solution of the full Maxwell equations---exploiting all aspects of wave physics arising in subwavelength scatterers---with inverse-scattering algorithms in a single large-scale optimization involving $\gtrsim 10^4$ degrees of freedom. The resulting structures scatter light in a way that is radically different from either a conventional lens or a random microstructure, and suppress the noise sensitivity of the inverse-scattering computation by several orders of magnitude. Incorporating the full wave physics is especially crucial for detecting spectral and polarization information that is discarded by geometric optics and scalar diffraction theory., Comment: 19 pages, 4 figures. Initially submitted on June 16, 2020. Expanded version submitted on October 10, 2020

We introduce single molecule light field microscopy (SMLFM), a new class of three-dimensional (3D) single molecule localization microscopy. By segmenting the back focal plane of a microscope objective with an array of microlenses to generate multiple 2D perspective views, the same single fluorophore can be imaged from different angles. These views, in combination with a bespoke fitting algorithm, enable the 3D positions of single fluorophores to be determined from parallax. SMLFM achieves up to 20 nm localization precision throughout an extended 6 µ m depth of field. The capabilities of SMLFM are showcased by imaging membranes of fixed eukaryotic cells and DNA nanostructures below the optical diffraction limit.

International audience; We introduce a new paradigm for solving regularized variational problems. These are typically formulated to address ill-posed inverse problems encountered in signal and image processing. The objective function is traditionally defined by adding a regularization function to a data fit term, which is subsequently minimized by using iterative optimization algorithms. Recently, several works have proposed to replace the operator related to the regularization by a more sophisticated denoiser. These approaches, known as plug-and-play (PnP) methods, have shown excellent performance. Although it has been noticed that, under nonexpansiveness assumptions on the denoisers, the convergence of the resulting algorithm is guaranteed, little is known about characterizing the asymptotically delivered solution. In the current article, we propose to address this limitation. More specifically, instead of employing a functional regularization, we perform an operator regularization, where a maximally monotone operator (MMO) is learned in a supervised manner. This formulation is flexible as it allows the solution to be characterized through a broad range of variational inequalities, and it includes convex regularizations as special cases. From an algorithmic standpoint, the proposed approach consists in replacing the resolvent of the MMO by a neural network (NN). We provide a universal approximation theorem proving that nonexpansive NNs provide suitable models for the resolvent of a wide class of MMOs. The proposed approach thus provides a sound theoretical framework for analyzing the asymptotic behavior of first-order PnP algorithms. In addition, we propose a numerical strategy to train NNs corresponding to resolvents of MMOs. We apply our approach to image restoration problems and demonstrate its validity in terms of both convergence and quality.

Through-focus scanning optical microscopy (TSOM) is one of the recommended measurement methods in semiconductor manufacturing industry in recent years because of its rapid and nondestructive properties. As a computational imaging method, TSOM takes full advantage of the information from defocused images rather than only concentrating on focused images. In order to improve the accuracy of TSOM in nanoscale dimensional measurement, this paper proposes a two-input deep-learning TSOM method based on Convolutional Neural Network (CNN). The TSOM image and the focused image are taken as the two inputs of the network. The TSOM image is processed by three columns convolutional channels and the focused image is processed by a single convolution channel for feature extraction. Then, the features extracted from the two kinds of images are merged and mapped to the measuring parameters for output. Our method makes effective use of the image information collected by TSOM system, for which the measurement process is fast and convenient with high accuracy. The MSE of the method can reach 5.18 nm2 in the measurement of gold lines with a linewidth range of 247–1010 nm and the measuring accuracy is much higher than other deep-learning TSOM methods.

We propose a real-time hyperspectral video acquisition system that uses coded slits. Conventional imaging spectrometers usually have scanning mechanisms that reduce the temporal resolution or sacrifice the spatial resolution to acquire spectral information instantly. Recently, computational spectral imaging has been applied to realize high-speed or high-performance spectral imaging. However, the most current computational spectral imaging systems take a long time to reconstruct spectral data cubes from limited measurements, which limits real-time hyperspectral video acquisition. In this work, we propose a new computational spectral imaging system. We substitute the slit in a conventional scanning-based imaging spectrometer with coded slits, which can achieve the parallel acquisition of spectral data and thus an imaging speed that is several times higher. We also apply an electronically controlled translation stage to use different codes at each exposure level. The larger amount of data allows for fast reconstruction through matrix inversion. To solve the problem of a trade-off between imaging speed and image quality in high-speed spectral imaging, we analyze the noise in the system. The severe readout noise in our system is suppressed with S-matrix coding. Finally, we build a practical prototype that can acquire hyperspectral video with a high spatial resolution and a high signal-to-noise ratio at 5 Hz in real time.

Point spread function (PSF) plays an important role in evaluating the quality of optical systems and imaging restoration. Conventional methods, which use a point-like source, acquire the PSF from the perspective of object points and suffer the problem of generating multiple point sources with sufficient intensity. Thus, in this study, we proposed a novel method to measure the PSF from the perspective of camera pixels. We considered spatially varying PSF as the light transport coefficients between the object and image points. By applying single-pixel imaging technology to every pixel of the image plane of a camera, PSF is extracted from the obtained light transport coefficients. Experiments also verified that the proposed method could achieve better signal-to-noise ratio PSF measurement than the conventional method and acquire an accurate blur model of the camera lens.

We propose a method of computational temporal ghost imaging over a single optical fiber using simple optical intensity detection instead of coherent detection. The transfer function of a temporal ghost imaging system over a single optical fiber is derived, which is a function of the total fiber dispersion and the power density spectrum of the light source. In the simulations, we find that the bandwidth of the transfer function will decrease rapidly along with the increase of fiber length or light source bandwidth, resulting in a significantly distorted ghost image. To improve the image quality, we first calculate the transfer function by approximating the optical spectrum of the light source based on intensity detection as its power density spectrum. Then, we use an optimized Fourier deconvolution to compute a high-quality image by deconvoluting the measured transfer function from the spectrum of the distorted image. Our experiments achieve single-arm ghost images of a 5-Gb/s nonreturn-to-zero pulse object over a 200-km fiber, which have almost the same quality as the two-arm ones.