Imaging Scenarios Research Articles

An Intelligent Cloud Storage Gateway for Medical Imaging.

Historically, medical imaging repositories have been supported by indoor infrastructures. However, the amount of diagnostic imaging procedures has continuously increased over the last decades, imposing several challenges associated with the storage volume, data redundancy and availability. Cloud platforms are focused on delivering hardware and software services over the Internet, becoming an appealing solution for repository outsourcing. Although this option may bring financial and technological benefits, it also presents new challenges. In medical imaging scenarios, communication latency is a critical issue that still hinders the adoption of this paradigm. This paper proposes an intelligent Cloud storage gateway that optimizes data access times. This is achieved through a new cache architecture that combines static rules and pattern recognition for eviction and prefetching. The evaluation results, obtained from experiments over a real-world dataset, show that cache hit ratios can reach around 80%, leading to reductions of image retrieval times by over 60%. The combined use of eviction and prefetching policies proposed can significantly reduce communication latency, even when using a small cache in comparison to the total size of the repository. Apart from the performance gains, the proposed system is capable of adjusting to specific workflows of different institutions.

Quantitation of tumor uptake with molecular breast imaging.

We developed scatter and attenuation-correction techniques for quantifying images obtained with Molecular Breast Imaging (MBI) systems. To investigate scatter correction, energy spectra of a 99m Tc point source were acquired with 0-7-cm-thick acrylic to simulate scatter between the detector heads. System-specific scatter correction factor, k, was calculated as a function of thickness using a dual energy window technique. To investigate attenuation correction, a 7-cm-thick rectangular phantom containing 99m Tc-water simulating breast tissue and fillable spheres simulating tumors was imaged. Six spheres 10-27 mm in diameter were imaged with sphere-to-background ratios (SBRs) of 3.5, 2.6, and 1.7 and located at depths of 0.5, 1.5, and 2.5 cm from the center of the water bath for 54 unique tumor scenarios (3 SBRs × 6 sphere sizes × 3 depths). Phantom images were also acquired in-air under scatter- and attenuation-free conditions, which provided ground truth counts. To estimate true counts, T, from each tumor, the geometric mean (GM) of the counts within a prescribed region of interest (ROI) from the two projection images was calculated as T=C1C2eμtF, where C are counts within the square ROI circumscribing each sphere on detectors 1 and 2, μ is the linear attenuation coefficient of water, t is detector separation, and the factor F accounts for background activity. Four unique F definitions-standard GM, background-subtraction GM, MIRD Primer 16 GM, and a novel "volumetric GM"-were investigated. Error in T was calculated as the percentage difference with respect to in-air. Quantitative accuracy using the different GM definitions was calculated as a function of SBR, depth, and sphere size. Sensitivity of quantitative accuracy to ROI size was investigated. We developed an MBI simulation to investigate the robustness of our corrections for various ellipsoidal tumor shapes and detector separations. Scatter correction factor k varied slightly (0.80-0.95) over a compressed breast thickness range of 6-9 cm. Corrected energy spectra recovered general characteristics of scatter-free spectra. Quantitatively, photopeak counts were recovered to <10% compared to in-air conditions after scatter correction. After GM attenuation correction, mean errors (95% confidence interval, CI) for all 54 imaging scenarios were 149% (-154% to +455%), -14.0% (-38.4% to +10.4%), 16.8% (-14.7% to +48.2%), and 2.0% (-14.3 to +18.3%) for the standard GM, background-subtraction GM, MIRD 16 GM, and volumetric GM, respectively. Volumetric GM was less sensitive to SBR and sphere size, while all GM methods were insensitive to sphere depth. Simulation results showed that Volumetric GM method produced a mean error within 5% over all compressed breast thicknesses (3-14 cm), and that the use of an estimated radius for nonspherical tumors increases the 95% CI to at most ±23%, compared with ±16% for spherical tumors. Using DEW scatter- and our Volumetric GM attenuation-correction methodology yielded accurate estimates of tumor counts in MBI over various tumor sizes, shapes, depths, background uptake, and compressed breast thicknesses. Accurate tumor uptake can be converted to radiotracer uptake concentration, allowing three patient-specific metrics to be calculated for quantifying absolute uptake and relative uptake change for assessment of treatment response.

Millimeter-wave spotlight imager using dynamic holographic metasurface antennas.

Computational imaging systems leverage generalized measurements to produce high-fidelity images, enabling novel and often lower cost hardware platforms at the expense of increased processing. However, obtaining full resolution images across a large field-of-view (FOV) can lead to slow reconstruction times, limiting system performance where faster frame rates are desired. In many imaging scenarios, the highest resolution is needed only in smaller subdomains of interest within a scene, suggesting an aperture supporting multiple modalities of image capture with different resolutions can provide a path to system optimization. We explore this concept in the context of millimeter-wave imaging, presenting the design and simulation of a single frequency (75 GHz), multistatic, holographic spotlight aperture integrated into a K-band (17.5-26.5 GHz), frequency-diverse imager. The spotlight aperture - synthesized using an array of dynamically tuned, holographic, metasurface antennas - illuminates a constrained region-of-interest (ROI) identified from a low-resolution image, extracting a high-fidelity image of the constrained-ROI with a minimum number of measurement modes. The designs of both the static, frequency-diverse sub-aperture and the integrated dynamic spotlight aperture are evaluated using simulation techniques developed for large-scale synthetic apertures.

Combined Magnetomotive ultrasound, PET/CT, and MR imaging of 68Ga-labelled superparamagnetic iron oxide nanoparticles in rat sentinel lymph nodes in vivo

Current methods for intra-surgical guidance to localize metastases at cancer surgery are based on radioactive tracers that cause logistical challenges. We propose the use of a novel ultrasound-based method, magnetomotive ultrasound (MMUS) imaging that employ a nanoparticle-based contrast agent that also may be used for pre-operative PET/MRI imaging. Since MMUS is radiation free, this eliminates the dependence between pre- and intra-operative imaging and the radiation exposure for the surgical staff. This study investigates a hypothetical clinical scenario of pre-operative PET imaging, combined with intra-operative MMUS imaging, implemented in a sentinel lymph node (SLN) rat model. At one-hour post injection of 68Ga-labelled magnetic nanoparticles, six animals were imaged with combined PET/CT. After two or four days, the same animals were imaged with MMUS. In addition, ex-vivo MRI was used to evaluate the amount of nanoparticles in each single SLN. All SLNs were detectable by PET. Four out of six SLNs could be detected with MMUS, and for these MMUS and MRI measurements were in close agreement. The MRI measurements revealed that the two SLNs undetectable with MMUS contained the lowest nanoparticle concentrations. This study shows that MMUS can complement standard pre-operative imaging by providing bedside real-time images with high spatial resolution.

Sponsorship of the FIFA World Cupp, Shareholder Wealth, and the Impact of Corruption

The FIFA World Cup™ is not only one of the most popular sporting events in the world: it is also famous for the extraordinary amount of sponsorship fees it attracts. To make sense economically, these sponsorship expenses should promise future cash flows that would make the investment profitable, so that it could create value for the stockholders of FIFA’s respective commercial affiliates. Investigations of stockholder reactions – which ultimately result from announced FIFA sponsorship deals – are rarely found in the literature. This study analyses the return effects for the stockholders of FIFA World Cup™ Commercial Affiliates, while distinguishing between positive and negative image scenarios pertaining to FIFA. It is shown that during a corruption-based downturn of FIFA’s reputation, stock prices of its commercial affiliates decrease significantly, while positive returns can be observed during periods of reputational stability. Furthermore, official FIFA Partners who choose to extend existing sponsorship contracts benefit most from the FIFA World Cup™ in terms of driving stockholder wealth. In addition, price effects are most pronounced for companies not involved in high-tech industries as well as for commercial affiliates with a high relative profitability.

Design of a space-based infrared imaging interferometer

Present space-based optical imaging sensors are expensive. Launch costs are dictated by weight and size, and system design must take into account the low fault tolerance of a system that cannot be readily accessed once deployed. We describe the design and first prototype of the space-based infrared imaging interferometer (SIRII) that aims to mitigate several aspects of the cost challenge. SIRII is a six-element Fizeau interferometer intended to operate in the short-wave and midwave IR spectral regions over a 6×6 mrad field of view. The volume is smaller by a factor of three than a filled-aperture telescope with equivalent resolving power. The structure and primary optics are fabricated from light-weight space-qualified carbon fiber reinforced polymer; they are easy to replicate and inexpensive. The design is intended to permit one-time alignment during assembly, with no need for further adjustment once on orbit. A three-element prototype of the SIRII imager has been constructed with a unit telescope primary mirror diameter of 165 mm and edge-to-edge baseline of 540 mm. The optics, structure, and interferometric signal processing principles draw on experience developed in ground-based astronomical applications designed to yield the highest sensitivity and resolution with cost-effective optical solutions. The initial motivation for the development of SIRII was the long-term collection of technical intelligence from geosynchronous orbit, but the scalable nature of the design will likely make it suitable for a range of IR imaging scenarios.

Incidental focal uptake in the breast and axilla on FDG PET: Clinical considerations and differential diagnosis

Incidental focal FDG uptake in the breast or axilla on PET/CT performed for evaluation of extra-mammary primary disease presents a diagnostic challenge. Radiologists must consider a broad differential diagnosis, assess clinical history, and judiciously employ other imaging modalities such as mammography, ultrasound and MRI in the pursuit of findings which help narrow the differential diagnosis. Tissue sampling may be reserved for nondiagnostic imaging scenarios.

Large size single image fast defogging and the real time video defogging FPGA architecture

Since most image defogging methods require complex image filtering computation, they cannot work effectively under large size real time image and video scenarios. In this work, targeting this problem, a large size single image computation acceleration bilateral filtering based defogging (CABFD) scheme is presented. And aiming for large size video real time defogging on smart TV, a simplified filtering alternative and the corresponding FPGA defogging architecture are also introduced. The experiments and the comparisons show that the proposed large size image fast defogging approach is effective and practical. And the proposed FPGA defogging architecture can be directly applied to realize Full HD (1920 × 1080) video displaying with real time synchronous defogging.

Spectral and temporal multiplexing for multispectral fluorescence and reflectance imaging using two color sensors.

Fluorescence imaging can reveal functional, anatomical or pathological features of high interest in medical interventions. We present a novel method to record and display in video rate multispectral color and fluorescence images over the visible and near infrared range. The fast acquisition in multiple channels is achieved through a combination of spectral and temporal multiplexing in a system with two standard color sensors. Accurate color reproduction and high fluorescence unmixing performance are experimentally demonstrated with a prototype system in a challenging imaging scenario. Through spectral simulation and optimization we show that the system is sensitive to all dyes emitting in the visible and near infrared region without changing filters and that the SNR of multiple unmixed components can be kept high if parameters are chosen well. We propose a sensitive per-pixel metric of unmixing quality in a single image based on noise propagation and present a method to visualize the high-dimensional data in a 2D graph, where up to three fluorescent components can be distinguished and segmented.

EIT Imaging of admittivities with a D-bar method and spatial prior: experimental results for absolute and difference imaging

Electrical impedance tomography (EIT) is an emerging imaging modality that uses harmless electrical measurements taken on electrodes at a body’s surface to recover information about the internal electrical conductivity and or permittivity. The image reconstruction task of EIT is a highly nonlinear inverse problem that is sensitive to noise and modeling errors making the image reconstruction task challenging. D-bar methods solve the nonlinear problem directly, bypassing the need for detailed and time-intensive forward models, to provide absolute (static) as well as time-difference EIT images. Coupling the D-bar methodology with the inclusion of high confidence a priori data results in a noise-robust regularized image reconstruction method. In this work, the a priori D-bar method for complex admittivities is demonstrated effective on experimental tank data for absolute imaging for the first time. Additionally, the method is adjusted for, and tested on, time-difference imaging scenarios. The ability of the method to be used for conductivity, permittivity, absolute as well as time-difference imaging provides the user with great flexibility without a high computational cost.

A new CT reconstruction technique using adaptive deformation recovery and intensity correction (ADRIC).

Sequential same-patient CT images may involve deformation-induced and non-deformation-induced voxel intensity changes. An adaptive deformation recovery and intensity correction (ADRIC) technique was developed to improve the CT reconstruction accuracy, and to separate deformation from non-deformation-induced voxel intensity changes between sequential CT images. ADRIC views the new CT volume as a deformation of a prior high-quality CT volume, but with additional non-deformation-induced voxel intensity changes. ADRIC first applies the 2D-3D deformation technique to recover the deformation field between the prior CT volume and the new, to-be-reconstructed CT volume. Using the deformation-recovered new CT volume, ADRIC further corrects the non-deformation-induced voxel intensity changes with an updated algebraic reconstruction technique ("ART-dTV"). The resulting intensity-corrected new CT volume is subsequently fed back into the 2D-3D deformation process to further correct the residual deformation errors, which forms an iterative loop. By ADRIC, the deformation field and the non-deformation voxel intensity corrections are optimized separately and alternately to reconstruct the final CT. CT myocardial perfusion imaging scenarios were employed to evaluate the efficacy of ADRIC, using both simulated data of the extended-cardiac-torso (XCAT) digital phantom and experimentally acquired porcine data. The reconstruction accuracy of the ADRIC technique was compared to the technique using ART-dTV alone, and to the technique using 2D-3D deformation alone. The relative error metric and the universal quality index metric are calculated between the images for quantitative analysis. The relative error is defined as the square root of the sum of squared voxel intensity differences between the reconstructed volume and the "ground-truth" volume, normalized by the square root of the sum of squared "ground-truth" voxel intensities. In addition to the XCAT and porcine studies, a physical lung phantom measurement study was also conducted. Water-filled balloons with various shapes/volumes and concentrations of iodinated contrasts were put inside the phantom to simulate both deformations and non-deformation-induced intensity changes for ADRIC reconstruction. The ADRIC-solved deformations and intensity changes from limited-view projections were compared to those of the "gold-standard" volumes reconstructed from fully sampled projections. For the XCAT simulation study, the relative errors of the reconstructed CT volume by the 2D-3D deformation technique, the ART-dTV technique, and the ADRIC technique were 14.64%, 19.21%, and 11.90% respectively, by using 20 projections for reconstruction. Using 60 projections for reconstruction reduced the relative errors to 12.33%, 11.04%, and 7.92% for the three techniques, respectively. For the porcine study, the corresponding results were 13.61%, 8.78%, and 6.80% by using 20 projections; and 12.14%, 6.91%, and 5.29% by using 60 projections. The ADRIC technique also demonstrated robustness to varying projection exposure levels. For the physical phantom study, the average DICE coefficient between the initial prior balloon volume and the new "gold-standard" balloon volumes was 0.460. ADRIC reconstruction by 21 projections increased the average DICE coefficient to 0.954. The ADRIC technique outperformed both the 2D-3D deformation technique and the ART-dTV technique in reconstruction accuracy. The alternately solved deformation field and non-deformation voxel intensity corrections can benefit multiple clinical applications, including tumor tracking, radiotherapy dose accumulation, and treatment outcome analysis.

Fullwave simulations of ultrasound propagation in the human body: Applications to imaging and motion estimation

The Fullwave simulation tool, which solves a modified Westervelt equation, can model 3D ultrasound propagation in the human body. It includes the effects of nonlinearity, frequency dependent attenuation, and scattering. This high order finite difference simulation has a high dynamic range and ability to include sub-resolution scattering physics enabling it to tackle the computationally challenging problem of generating ultrasound images directly from the first principles of propagation and reflection. Three dimensional acoustical maps of the human body are derived from the Visible Human project, which provides the high degree of anatomical fidelity required to model sources of image degradation such as reverberation clutter and aberration. A three dimensional intercostal imaging scenario shows how the ribs degrade the image quality via reverberation and yet can improve the beam shape via apodization. It is shown that the in situ Mechanical Index (MI) in the human body differs significantly from derated estimates in a homogeneous medium. Due to tissue heterogeneities, the peak MI is close to the transducer and far from the focal region. Finally, we present a numerical model of subresolution displacements by using an impedance flow model applied to a scatterer composed of two elements. This shows that the numerical model can support displacements that are over three orders of magnitude smaller than the grid spacing. Applications to acoustic radiation force, elastography, and shear shock wave tracking are discussed.

Technical Note: On the accuracy of parametric two-parameter photon cross-section models in dual-energy CT applications.

To evaluate and compare the theoretically achievable accuracy of two families of two-parameter photon cross-section models: basis vector model (BVM) and modified parametric fit model (mPFM). The modified PFM assumes that photoelectric absorption and scattering cross-sections can be accurately represented by power functions in effective atomic number and/or energy plus the Klein-Nishina cross-section, along with empirical corrections that enforce exact prediction of elemental cross-sections. Two mPFM variants were investigated: the widely used Torikoshi model (tPFM) and a more complex "VCU" variant (vPFM). For 43 standard soft and bony tissues and phantom materials, all consisting of elements with atomic number less than 20 (except iodine), we evaluated the theoretically achievable accuracy of tPFM and vPFM for predicting linear attenuation, photoelectric absorption, and energy-absorption coefficients, and we compared it to a previously investigated separable, linear two-parameter model, BVM. For an idealized dual-energy computed tomography (DECT) imaging scenario, the cross-section mapping process demonstrates that BVM more accurately predicts photon cross-sections of biological mixtures than either tPFM or vPFM. Maximum linear attenuation coefficient prediction errors were 15% and 5% for tPFM and BVM, respectively. The root-mean-square (RMS) prediction errors of total linear attenuation over the 20keV to 1000keV energy range of tPFM and BVM were 0.93% (tPFM) and 0.1% (BVM) for adipose tissue, 0.8% (tPFM) and 0.2% (BVM) for muscle tissue, and 1.6% (tPFM) and 0.2% (BVM) for cortical bone tissue. With exception of the thyroid and Teflon, the RMS error for photoelectric absorption and scattering coefficient was within 4% for the tPFM and 2% for the BVM. Neither model predicts the photon cross-sections of thyroid tissue accurately, exhibiting relative errors as large as 20%. For the energy-absorption coefficients prediction error, RMS errors for the BVM were less than 1.5%, while for the tPFM, the RMS errors were as large as 16%. Compared to modified PFMs, BVM shows superior potential to support dual-energy CT cross-section mapping. In addition, the linear, separable BVM can be more efficiently deployed by iterative model-based DECT image-reconstruction algorithms.

Sparsity-Driven SAR Imaging for Highly Maneuvering Ground Target by the Combination of Time-Frequency Analysis and Parametric Bayesian Learning

In this paper, a well-focused synthetic aperture radar (SAR) image for highly maneuvering ground target is formed. For high-resolution SAR imaging, the phase modulation from the maneuverer's high-order movements severely degrades the focusing quality of the target image, if the conventional SAR imaging algorithm under the constant target velocity assumption is used. To deal with this problem, a new SAR ground moving target imaging (GMTIm) algorithm is presented with a two-step strategy to obtain a high-resolution maneuvering target image with highly focused responses. Pseudo Wigner–Ville distribution is first employed to access and compensate for the bulk of the high-order phase. Then, to further enhance the target image quality, the SAR-GMTIm problem is solved by sparse Bayesian learning (SBL), where an accurate phase autofocusing is incorporated for the compensation of the residual high-order phase. A novel time-frequency representation, known as Lv's distribution, is adopted to determine the parametric dictionary used in the SBL processing. To accommodate the possible multiple-target imaging scenario, the intended SAR-GMTIm algorithm is developed in a coarse-to-fine compensation procedure. Finally, both simulated data and real Gotcha data are applied to validate the effectiveness and superiority of the proposed SAR imaging algorithm for ground maneuvering targets.

Interlaced zone plate optics for hard X-ray imaging in the 10\u2009nm range

Multi-keV X-ray microscopy has been particularly successful in bridging the resolution gap between optical and electron microscopy. However, resolutions below 20 nm are still considered challenging, as high throughput direct imaging methods are limited by the availability of suitable optical elements. In order to bridge this gap, we present a new type of Fresnel zone plate lenses aimed at the sub-20 and the sub-10 nm resolution range. By extending the concept of double-sided zone plate stacking, we demonstrate the doubling of the effective line density and thus the resolution and provide large aperture, singlechip optical devices with 15 and 7 nm smallest zone widths. The detailed characterization of these lenses shows excellent optical properties with focal spots down to 7.8 nm. Beyond wave front characterization, the zone plates also excel in typical imaging scenarios, verifying their resolution close to their diffraction limited optical performance.

"Choosing Wisely" Imaging Recommendations: Initial Implementation in New England Emergency Departments.

IntroductionIn June 2016, the American College of Emergency Physicians (ACEP) Emergency Quality Network began its Reduce Avoidable Imaging Initiative, designed to “reduce testing and imaging with low risk patients through the implementation of Choosing Wisely recommendations.” However, it is unknown whether New England emergency departments (ED) have already implemented evidence-based interventions to improve adherence to ACEP Choosing Wisely recommendations related to imaging after their initial release in 2013. Our objective was to determine this, as well as whether provider-specific audit and feedback for imaging had been implemented in these EDs.MethodsThis survey study was exempt from institutional review board review. In 2015, we mailed surveys to 195 hospital-affiliated EDs in all six New England states to determine whether they had implemented Choosing Wisely-focused interventions in 2014. Initial mailings included cover letters denoting the endorsement of each state’s ACEP chapter, and we followed up twice with repeat mailings to non-responders. Data analysis included descriptive statistics and a comparison of state differences using Fisher’s exact test.ResultsA total of 169/195 (87%) of New England EDs responded, with all individual state response rates >80%. Overall, 101 (60%) of responding EDs had implemented an intervention for at least one Choosing Wisely imaging scenario; 57% reported implementing a specific guideline/policy/clinical pathway and 28% reported implementing a computerized decision support system. The most common interventions were for chest computed tomography (CT) in patients at low risk of pulmonary embolism (47% of EDs) and head CT in patients with minor trauma (45% of EDs). In addition, 40% of EDs had implemented provider-specific audit and feedback, without significant interstate variation (range: 29–55%).ConclusionOne year after release of the ACEP Choosing Wisely recommendations, most New England EDs had a guideline/policy/clinical pathway related to at least one of the recommendations. However, only a minority of them were using provider-specific audit and feedback or computerized decision support. Few EDs have embraced the opportunity to implement the multiple evidence-based interventions likely to advance the national goals of improving patient-centered and resource-efficient care.

Closed-form expressions for flip angle variation that maximize total signal in T1-weighted rapid gradient echo MRI.

Magnetization-prepared rapid gradient-echo (MPRAGE) sequences are commonly employed for T1-weighted structural brain imaging. Following a contrast preparation radiofrequency (RF) pulse, the data acquisition proceeds under nonequilibrium conditions of the relaxing longitudinal magnetization. Variation of the flip angle can be used to maximize total available signal. Simulated annealing or greedy algorithms have so far been published to numerically solve this problem, with signal-to-noise ratios optimized for clinical imaging scenarios by adhering to a predefined shape of the signal evolution. We propose an unconstrained optimization of the MPRAGE experiment thatemploys techniques from resource allocation theory. A new dynamic programming solution is introduced that yields closed-form expressions for optimal flip angle variation. Flip angle series are proposed that maximize total transverse magnetization (Mxy) for a range of physiologic T1 values. A 3D MPRAGE sequence is modified to allow for a controlled variation of the excitation angle. Experiments employing a T1 contrast phantom are performed at 3T. 1D acquisitions without phase encoding permit measurement of the temporal development of Mxy. Image mean signal and standard deviation for reference flip angle trains are compared in 2D measurements. Signal profiles at sharp phantom edges are acquired to access image blurring related to nonuniform Mxy development. A novel closed-form expression for flip angle variation is found that constitutes the optimal policy to reach maximum total signal. It numerically equals previously published results of other authors when evaluated under their simplifying assumptions. Longitudinal magnetization (Mz) is exhaustively used without causing abrupt changes in the measured MR signal, which is a prerequisite for artifact free images. Phantom experiments at 3T verify the expected benefit for total accumulated k-space signal when compared with published flip angle series. Describing the MR signal collection in MPRAGE sequences as a Bellman problem is a new concept. By means of recursively solving a series of overlapping subproblems, this leads to an elegant solution for the problem of maximizing total available MR signal in k-space. A closed-form expression for flip angle variation avoids the complexity of numerical optimization and eases access to controlled variation in an attempt to identify potential clinical applications.

Efficient 3D morphable face model fitting

3D face reconstruction of shape and skin texture from a single 2D image can be performed using a 3D Morphable Model (3DMM) in an analysis-by-synthesis approach. However, performing this reconstruction (fitting) efficiently and accurately in a general imaging scenario is a challenge. Such a scenario would involve a perspective camera to describe the geometric projection from 3D to 2D, and the Phong model to characterise illumination. Under these imaging assumptions the reconstruction problem is nonlinear and, consequently, computationally very demanding. In this work, we present an efficient stepwise 3DMM-to-2D image-fitting procedure, which sequentially optimises the pose, shape, light direction, light strength and skin texture parameters in separate steps. By linearising each step of the fitting process we derive closed-form solutions for the recovery of the respective parameters, leading to efficient fitting. The proposed optimisation process involves all the pixels of the input image, rather than randomly selected subsets, which enhances the accuracy of the fitting. It is referred to as Efficient Stepwise Optimisation (ESO).The proposed fitting strategy is evaluated using reconstruction error as a performance measure. In addition, we demonstrate its merits in the context of a 3D-assisted 2D face recognition system which detects landmarks automatically and extracts both holistic and local features using a 3DMM. This contrasts with most other methods which only report results that use manual face landmarking to initialise the fitting. Our method is tested on the public CMU-PIE and Multi-PIE face databases, as well as one internal database. The experimental results show that the face reconstruction using ESO is significantly faster, and its accuracy is at least as good as that achieved by the existing 3DMM fitting algorithms. A face recognition system integrating ESO to provide a pose and illumination invariant solution compares favourably with other state-of-the-art methods. In particular, it outperforms deep learning methods when tested on the Multi-PIE database.

Atomic vapor as a source of tunable, non-Gaussian self-reconstructing optical modes

In this manuscript, we demonstrate the ability of nonlinear light-atom interactions to produce tunably non-Gaussian, partially self-healing optical modes. Gaussian spatial-mode light tuned near to the atomic resonances in hot rubidium vapor is shown to result in non-Gaussian output mode structures that may be controlled by varying either the input beam power or the temperature of the atomic vapor. We show that the output modes exhibit a degree of self-reconstruction after encountering an obstruction in the beam path. The resultant modes are similar to truncated Bessel-Gauss modes that exhibit the ability to self-reconstruct earlier upon propagation than Gaussian modes. The ability to generate tunable, self-reconstructing beams has potential applications to a variety of imaging and communication scenarios.

Three‐dimensional bistatic array imaging using range migration algorithm

Millimetre wave imaging systems are widely used in many fields, such as non-destructive testing, security screening and medical imaging. For security-related personnel screening, the geometries of planar and cylindrical-scanning antenna arrays are often adopted. A planar array imager is capable of acquiring a three-dimensional image in real time, but the information on the back side of the object will be lost. A cylindrical scanning imager is well suited for constructing an image in full 360° directions, what is lost is the time that it takes to complete a full scan. Regardless, mechanical scanning geometry is not always the best approach for real-time systems. In this letter, an imaging scenario is proposed, which is composed of two parallel arrays facing each other. Multiple input multiple output technology is applied to the pair of arrays. Range migration algorithm is adapted specially for this scenario, through which nearly 360°-images of high quality can be constructed in real time when an object is placed between the two arrays.