External Detector Research Articles

Quasiprobabilities in Quantum Thermodynamics and Many-Body Systems

In this tutorial, we present the definition, interpretation, and properties of some of the main quasiprobabilities that can describe the statistics of measurement outcomes evaluated at two or more times. Such statistics incorporate the incompatibility of the measurement observables and the state of the measured quantum system. We particularly focus on Kirkwood-Dirac quasiprobabilities and related distributions. We also discuss techniques to experimentally access a quasiprobability distribution, ranging from the weak two-point measurement scheme, to a Ramsey-like interferometric scheme and procedures assisted by an external detector. Once defined the fundamental concepts following the standpoint of joint measurability in quantum mechanics, we illustrate the use of quasiprobabilities in quantum thermodynamics to describe the quantum statistics of work and heat, and to explain anomalies in the energy exchanges entailed by a given thermodynamic transformation. On the one hand, in work protocols, we show how absorbed energy can be converted to extractable work and vice versa due to Hamiltonian incompatibility at distinct times. On the other hand, in exchange processes between two quantum systems initially at different temperatures, we explain how quantum correlations in their initial state may induce cold-to-hot energy exchanges, which are unnatural between any pair of equilibrium nondriven systems. We conclude the tutorial by giving simple examples where quasiprobabilities are applied to many-body systems: scrambling of quantum information, sensitivity to local perturbations, and quantum work statistics in the quenched dynamics of models that can be mapped onto systems of free fermions, for instance, the Ising model with a transverse field. Throughout the tutorial, we meticulously present derivations of essential concepts alongside straightforward examples, aiming to enhance comprehension and facilitate learning. Published by the American Physical Society 2024

Apatite Fission-Track Dating: A Comparative Study of Ages Obtained by the Automated Counting LA-ICP-MS and External Detector Methodologies

Abstract Laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) in combination with developments in digital microscopy, image analysis, and computer software has allowed the implementation of an automated counting approach for the apatite fission-track (AFT) analysis. We refer to this approach as the “automated counting-LA-ICP-MS” (ACLA) method. Two major components comprise the ACLA method: (i) the digital counting of spontaneous tracks performed on high-resolution images captured from apatite grains and (ii) the measurement of 238U content in apatite by LA-ICP-MS. This study includes ACLA analyses from Fish Canyon Tuff (FCT) and Durango apatite standard crystals. Furthermore, a comparative age study between the ACLA and conventional external detector method (EDM) strategies was performed on a set of thirteen granitoid samples from northwestern Mexico and four granitic samples from the eastern Dharwar craton (EDC), India. ACLA analyses on FCT yielded an AFT age of 28.1 ± 0.6 (1σ) and 28.8 ± 1.1 (1σ) Ma for Durango apatite, whereas reported EDM ages are 27.5 ± 0.5 and 31.4 ± 0.5 Ma, respectively. Calculated AFT ages using the ACLA method from northwestern Mexico samples range from 11.1 ± 1.1 to 42.0 ± 3.6 Ma (EDM ages range from 10.0 ± 0.8 to 54.0 ± 3.0 Ma), whereas AFT ages from the EDC samples range from 147 ± 3.1 to 220.5 ± 12.5 Ma (EDM ages range from 120.9 ± 4.5 to 197.1 ± 19.4 Ma). Based on a statistical comparison with ages previously determined by the conventional EDM on the same samples and considering their 2σ uncertainties, these ages are in good agreement.

Optical proximity sensors using multiple quantum well didoes

InGaN/GaN multiple quantum well (MQW) diodes perform multiple functions, such as optical emission, modulation and reception. In particular, the partially overlapping spectral region between the electroluminescence (EL) and responsivity spectra of each diode results in each diode being able to sense light from another diode of the same MQW structure. Here, we present a noncontact, optical proximity sensing system by integrating an MQW-based light transmitter and detector into a tiny GaN-on-sapphire chip. Changes in the external environment modulate the light emitted from the transmitter. Reflected light is received by the on-chip MQW detector, wherein the carried external modulation information is converted into electrical signals that can be extracted. The maximum detection proximity is approximately 17 mm, and the displacement detection accuracy is within 1 mm. Based on the detection of distance, we extend the application of the sensor to vibration and pressure detection. This monolithic integration design can replace external discrete light transmitter and detector systems to miniaturize reflective sensor architectures, enabling the development of novel optical sensors.

WaveGate: a versatile tool for temporal shaping of synchrotron beams.

We present a full performance characterization of a solid state pulse picker for hard x-ray pulses at synchrotrons. The device is called WaveGate. Specifically, we quantify its efficiency (>30 %), timing capabilities (switching times between 100 ns and ms), on-off contrast (>104) and influence on the coherence properties of the incident x-ray beam. In addition, we discuss the implementation of the WaveGate in an optical pump - x-ray probe setup. Even if single pulse selection is performed by external detector gating, the WaveGate drastically increases the efficiency of a measurement. Finally, we introduce advanced timing schemes that can be realized by modulating the time structure of the synchrotron beam.

Fission-track age calibration: Phi and Zeta, never the twain shall meet?

The fission-track method was intended to be an autonomous dating method requiring two track counts and a neutron fluence measurement. It became clear that it was beset with methodological problems, and neutron fluence measurements were abandoned in favour of age standards. The coming of mass-spectrometric uranium measurements eliminated the need for irradiations altogether, and stimulated renewed efforts to establish fission-track dating as an autonomous method. This ran into similar problems as before, and recent publications signal a return to age standards. The attempts at independent dating nevertheless revealed that experts agree on the recommended fission constant and the need to consider experimental factors related to the track counts. In spite of that, the renewed reliance on standards makes that fission-track ages are again not independent, but relative, ages with an uncertain meaning, and from their nature untestable.We report an attempt at standardless and standard-based dating of the Durango and Duluth apatites using neutron irradiation and the external detector method. We start from the assumption that neutron fluence measurements are the most precise and accurate achievable in fission-track dating, a consensus on the recommended fission constant, and on the need to consider experimental factors. One factor (ζ0) accounts for the combined experimental influences; ζ0 can be calculated from identifiable contributions related to track registration, etching and counting; ζ0 can also be determined by experiment and calculated from the ages of standards. This gives three separate, consistent, dimensionless calibration factors that, in contrast to Z and ζ, are independent of the neutron fluence or the uranium content of a reference glass; ζ0 remains a personal factor, however.The standardless fission-track ages of the Durango age standard are in agreement with its reference age and the standardless and standard-based ages of the Duluth apatites are consistent with each other. Furthermore, if ζ0 is used to correct conventional Z- (and by implication ζ-) values for experimental effects, the results are consistent with their definitions in terms of physical parameters. Although ζ0 is a personal factor, the small number of measurements to date is consistent with each other. In contrast, annealing corrections based on the fossil and induced confined fission track lengths overestimate the reference age of Durango and produce incompatible standardless and standard-based ages of the dated apatites. We argue that this points to unresolved issues with the relationship between the age and the fossil track length, and suggest a connection with etching.Plots of the single-grain ages against uranium content and radial plots of the data exhibit conspicuous structure. However statistical factors alone account for the nature and magnitude of the observed correlations. This implies that in the case of the Duluth apatites there is no identifiable effect of radiation damage on the single-grain ages despite their high accumulated α-doses and protracted thermal histories. Spurious correlations must be addressed before attempting to interpret the observed data structure in terms of radiation damage, apatite composition or age components.

Biophysical approach to modeling reflection: basis, methods, results

The approach used by physics is based on the identification and study of ideal objects, which is also the basis of biophysics, in combination with von Neumann heuristic modeling and functional fractionation according to R.Rosen is discussed as a tool for studying the properties of consciousness. The object of the study is a kind of line of analog systems: the human brain, the vertebrate brain, the invertebrate brain and artificial neural networks capable of reflection, which is a key property characteristic of consciousness. Reflection in the broad sense of the word, understood as an internal representation of the external world, is characteristic of a wide range of animals, and some of them (bumblebees, fish) even demonstrate reflection in the narrow sense of the word, understood as an inner self-representation. This complex behavior is realized by miniature brains of ~1 million neurons. The use of simple recurrent neural networks (RNNs) to obtain answers to general questions is illustrated. For example, it has been shown a small RNS is able to pass delayed matching to sample (DMTS) test, forming an individual dynamic representation of the received stimulus, allowing decoding by a special external neural detector. . It has been demonstrated in the reflexive game “even-odd”, the RNS has a huge advantage over a multi-layered neural network, with the same and a larger number of neurons – reflection defeats regression. It was found that the asymmetry of outcomes in the odd-even game, which was explained by various causes, including psychological ones – “it’s easier to catch up than to run away”, is reproduced in the game of two RNNs. Obviously, there are no psychological causes here and the advantage of the player playing for “even” is explained by the more complex strategy of the “odd” player – he needs to predict the opponent’s move and choose the opposite one.

A dentist’s perspective on scintigraphy about its applications as a diagnostic tool for a myriad of diseases affecting oral health

A "branch or specialty of medicine & medical imaging that uses radionuclides and relies on the process of radioactive decay in the diagnosis and treatment of disease" is known as nuclear medicine. Assessing physiologic change, which follows biochemical changes, is made easier with the use of radionuclide imaging, which is sometimes referred to as a functional imaging approach. Scintigraphy is a diagnostic procedure used in nuclear medicine where radioisotopes are administered internally into the body in liquid or gaseous forms, and the distinctive radiation that emerges is captured by external detectors known as gamma cameras, producing two-dimensional pictures. Bone scintigraphy, lymphoscintigraphy, salivary gland scintigraphy, and radio-immuno-scintigraphy are a few examples of scintigraphy techniques.Nuclear imaging has the advantage of giving extremely high levels of diagnostic sensitivity, which is successful in identifying even the smallest pathophysiological changes that are highlighted in the process of diagnosing early-stage to progressive diseases, making it the preferred diagnostic modality.

A simple and rapid external standard calibration HPLC method for determination of lumefantrine in dried blood spot samples from malaria patients in Botswana.

A simple external calibration liquid chromatography-diode array detector method was developed, validated, and applied for the determination of lumefantrine (Lum) in dried blood spot (DBS) samples collected from malaria patients in Botswana. The samples were validated in accordance with the United States Food and Drug Administration guidelines for bioanalytical methods after sample preparation using solid-liquid extraction. Separation was achieved using an XTerra C18 column (50 × 4.6mm,5μm), and a binary solvent system of acetonitrile and water adjusted to pH2.3 was used as the mobile phase. The validated method was applied for the determination of Lum in DBS samples collected from malaria patients infected with Plasmodium falciparum in Botswana. The calibration curve was linear between 0.5 and 12 μg/mL with a coefficient of determination (R2 ) of 0.9996. The limit of detection and the lower limit of quantification were 0.5 and 1.4μg/mL, respectively. The efficiency of extraction measured as percentage recovery ranged between 84.2% and 107.8% at the three quality control (QC) levels, that is, low QC, mid QC, and high QC. In conclusion, data suggest that the method is suitable for the determination of trace Lum in biofluids and can also be used for therapeutic drug monitoring and pharmacokinetic profiling.

Test-beam performance results of the FASTPIX sub-nanosecond CMOS pixel sensor demonstrator

Within the ATTRACT FASTPIX project, a monolithic pixel sensor demonstrator chip was developed in a modified 180 nm CMOS imaging process technology, targeting sub-nanosecond timing precision for single ionizing particles. It features a small collection electrode design on a 25 μm-thick epitaxial layer and contains 32 mini matrices of 68 hexagonal pixels each, with pixel pitches ranging from 8.66 to 20 μm. Four pixels are transmitting an analog output signal and 64 are transmitting binary hit information. Various design variations are explored, aiming at accelerating the charge collection and making the timing of the charge collection more uniform over the pixel area. Signal treatment of the analog waveforms, as well as reconstruction of digital position, time and charge information, is carried out off-chip. This contribution introduces the design of the sensor and readout system and presents performance results for various pixel designs achieved in recent test-beam measurements with external tracking and timing reference detectors. A time resolution below 150 ps is obtained at full efficiency for all pixel pitches.

Ultrathin Pyroelectric Photodetector with Integrated Polarization-Sensing Metasurface.

An abundance of metallic metasurfaces have been realized with miniscule, intricate features capable of tailored scattering, reflection, and absorption; however, high losses through heat limit their use in optoelectronics. Here, codesign of a detector and a polarization-sensing metasurface overcomes this challenge by utilizing the heat generation for integrated pyroelectric detection of the incoming light polarization. Using a nanogap metasurface with asymmetric metallic elements, polarization-sensitive photodetection exhibits high extinction ratios up to 19 for orthogonally polarized light and allows extraction of Stokes parameters with <12% deviation from theoretical values. This polarization-sensitive photodetector is ultrathin, consisting of active layers of only 290 nm, and exhibits fast response times of ∼2 ns. The structure is fully integrated, requiring no external cameras, detectors, or power sources, and points toward the creation of layered, multifunctional devices that utilize exotic metasurface properties for novel and compact sensing and imaging.

LA-ICP-MS APATITE FISSION-TRACK DATING OF THE SIBERIAN TRAPS INTRUSIONS: METHOD, FIRST RESULTS AND THEIR INTERPRETATION

In this paper we present the results on apatite fission-track dating, which was first performed at the Schmidt Institute of Physics of the Earth, Russian Academy of Sciences, using the laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS, Dobretsov Geological Institute, Siberian Branch of the Russian Academy of Sciences) for two intrusions of the Siberian Permian-Triassic large igneous province: Magan alkaline-ultramafic pluton (two samples) and Kontay intrusion (one sample). The obtained AFT ages are 217.6±18.6 and 238.8±35.8 Ma (95% confidence) for the Magan intrusion and 150.0±23.0 Ma for the Kontay intrusion and mark the time since they have been cooled below 120 °C. The distributions of track lengths in apatite grains from the studied samples indicate their rapid cooling to near-surface temperatures. We provide a detailed description of the method used, and also demonstrate that the results of fissiontrack analysis performed on the "sample-to-sample" principle by the classical external detector method (EDM) and the LA-ICP-MS method in the modification of the zeta calibration coincide within the age uncertainty.

Part-Based Representation Enhancement for Occluded Person Re-Identification

Retrieving an occluded pedestrian remains a challenging problem in person re-identification (re-id). Most existing methods utilize external detectors to disentangle the visible body parts. However, these methods are unstable due to domain bias and consume numerous computing resources. In this paper, we propose a novel and lightweight Part-based Representation Enhancement (PRE) network for occluded re-id that takes full advantages of the local correlations to aggregate distinctive information for local features without relying on auxiliary detectors. First, according to the information qualities of different body parts, we design a reasonable partition strategy to obtain the local features. Next, a Partial Relationship Aggregation (PRA) module is developed to self-mine the visibility of the body and construct a correlation matrix for collecting the information related to pre-defined classes. Following this, we propose an Inter-part Omnibearing Fusion (IOF) module that leverages the occlusion-suppressed class features to enhance the distinctiveness of the local features via feature completion and reverse fusion strategies. During the testing phase, the global and reconstructed local features are concatenated together for re-id without a complex visible region matching algorithm. Extensive experiments on occluded, partial, and holistic re-id benchmarks demonstrate the superiority of PRE over state-of-the-art methods in terms of accuracy and model complexity.

A grain-by-grain comparison of apatite fission-track analysis by LA-ICP-MS and the External Detector Method

Laser-ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) is increasingly used in fission-track analysis to determine the uranium content of host mineral specimens, particularly apatite. Fission-track dating by LA-ICP-MS (LAFT) has several advantages over the conventional External Detector Method (EDM), particularly in terms of sample turn-around times and the fact that neutron irradiations and the handling of radioactive materials are no longer necessary, while providing a similar level of in-situ information about parent nuclide (238U) concentrations. In addition, it facilitates the simultaneous measurement of multiple isotopes for double or triple-dating approaches or compositional characterisation. While it is often implicitly assumed that the EDM and LAFT fission-track dating approaches produce equivalent results, this assertion has yet to be adequately tested. We present an extensive dataset of apatite fission track results from 17 samples representing a large range of fission-track ages (∼0–2 Ga), 238U concentrations (0.14–410 ppm) and thermal histories that were analysed grain-by-grain using both techniques in order to investigate whether they yield concordant results during routine fission-track analysis.Apart from a few outliers, our data show that 238U concentrations measured by the EDM and LAFT techniques yield indistinguishable results across at least three orders of magnitude when a similar calibration system against rapidly cooled standards (e.g., Durango) is used. Comparison of single grain pooled and central ages reveals that LAFT ages are within error of EDM ages for apatite fission track standards such as Fish Canyon Tuff or Durango, as well as for a range of other samples whose shorter mean confined track lengths (<13 μm) and broader track distributions indicate they experienced more complex cooling histories. The most important conclusion here is that both the conventional EDM and LAFT methods can be expected to yield identical results for the breadth of ages, 238U concentrations, and underlying thermal histories commonly found in real world apatites.Importantly, the aggregate empirical calibrations for EDM and LAFT mask an underlying assumption that the mean etchable range of fission fragments is a constant having the mean value observed for spontaneous tracks in age standards such as the Durango apatite. Given that this assumption is known to be false in the great majority of samples, it is our view that empirically derived EDM and LAFT fission-track ages are best considered as model ages and that there should be greater clarity about the assumptions involved in their calculation.

Residence Time Distribution Measurements and Modeling in an Industrial-Scale Siemens Flotation Cell

This short communication presents residence time distribution (RTD) measurements and modeling in a 16 m3 Siemens flotation cell, as the first RTD characterization in an industrial-scale pneumatic cell. The Siemens cell was installed as a pre-rougher machine in a Cu-Mo selective plant. This plant recovered molybdenite as an enriched product, depressing copper-bearing minerals. Irradiated non-floatable solid and Br82 in water solution were employed as solid and liquid tracers, respectively. The tracers were instantly injected into the Siemens cell, and the inlet and outlet concentrations were directly measured by external non-invasive detectors. From the flotation literature, three model structures for the RTDs were evaluated, including perfect mixing, one large perfect mixer and one small perfect mixer in series (LSTS), and N perfectly mixed reactors in series. A transport delay was incorporated for all models. The LSTS representation was more consistent with the experimental data, showing that the Siemens cell RTDs presented significant deviations with respect to perfect mixing and plug-flow regimes. From the industrial measurements, mean residence times of 4.1–5.2 min were estimated.

Full-BAPose: Bottom Up Framework for Full Body Pose Estimation

We present Full-BAPose, a novel bottom-up approach for full body pose estimation that achieves state-of-the-art results without relying on external people detectors. The Full-BAPose method addresses the broader task of full body pose estimation including hands, feet, and facial landmarks. Our deep learning architecture is end-to-end trainable based on an encoder-decoder configuration with HRNet backbone and multi-scale representations using a disentangled waterfall atrous spatial pooling module. The disentangled waterfall module leverages the efficiency of progressive filtering, while maintaining multi-scale fields-of-view comparable to spatial pyramid configurations. Additionally, it combines multi-scale features obtained from the waterfall flow with the person-detection capability of the disentangled adaptive regression and incorporates adaptive convolutions to infer keypoints more precisely in crowded scenes. Full-BAPose achieves state-of-the art performance on the challenging CrowdPose and COCO-WholeBody datasets, with AP of 72.2% and 68.4%, respectively, based on 133 keypoints. Our results demonstrate that Full-BAPose is efficient and robust when operating under a variety conditions, including multiple people, changes in scale, and occlusions.

Insights into handling and delivery of Y-90 radioembolization therapies.

The use of Y-90 radioembolization techniques has become a standard tool for the treatment of liver cancer and metastatic diseases that result in liver lesions. As there are only two approved forms of radioembolization therapy, the procedures for use are also fairly standardized even though exact international and interdepartmental procedures can vary. What has been less published over the years are the nuanced differences in delivery techniques and handling of the two available Y90 radioembolization therapies. This paper seeks to examine various aspects of delivery techniques, product handling, and radiation exposure that differ between the available and approved products. Understanding these differences can assist with providing more efficient treatment, confirmation of accurate therapy, more informed handling of the products, and improved training of physicians and other hospital staff. Two commercially available and approved radioembolization devices were compared to assess nuanced, but key differences between the available products regarding therapy delivery, handling of the products, and radiation exposure to patients and staff. This work is broken into two sections: (1) Therapy Delivery, (2) Radiation Safety. Therapy delivery characteristics were assessed by using an external radiation detector system with detectors placed inside of each delivery system facing the dose vial and on the output catheter lines to the patient. Additional detectors were placed near the liver of the patient and on top of the foot to measure extremities. Data were acquired continuously throughout therapy delivery to collect time activity curves (TACs) for the characterization of each therapy. These data were analyzed to assess if (a) real-time monitoring of radiation could be used to provide an accurate assessment of residual dose before the patient leaves the procedure room, and (b) can dose delivery characteristics be observed that enable improved training and quality control. Calculation of residual dose using the external detector TACs was performed by analyzing initial and final activity peaks to determine measured count rate differences. Radiation safety aspects were assessed by monitoring radiation exposure to staff handling each of the available therapy products. Nuclear medicine technologists and interventional radiology physician body and hand doses were measured for each delivered therapy using standard body and ring dosimeters. The TACs noted above collected for the liver and extremities were used to assess if any off-target or leached Y90 activity could be detected for each therapy. Blood was collected at times before, during, and after treatment and then counted on a gamma counter to assess differences in free Y90 circulating in the blood. Each patient in this study also received a post-treatment whole-body PET/CT at 2-4 h post-infusion to assess for any aggregate free Y90 deposition that may have resulted from circulating free Y90 in the subject following therapy. Calculations of residual dose in the vial following therapy using the real-time detection methods resulted in values that were not statistically different from the values calculated by nuclear medicine following the procedure ( ). Real-time collection of dose delivery data enabled observation of key characteristics related to each delivery method. For SIR-spheres procedures, the cycle of pushing the dose and visualizing with fluoro can easily be seen with each push resulting in a smaller and smaller peak with intermittent fluoroscopy pulses. TheraSpheres infusions show a rapid bolus with nearly all of the measurable injected activity being infused in the first push of the dose. Staff radiation exposure assessments showed statistically significant differences between glass and resin spheres for hand doses of physicians and technologists (p > 0.05), but no statistical difference between body doses for both products ( ). Assessments of free Y90 circulating during therapy showed that patients undergoing therapies with resin spheres had post-infusion blood levels that were 120% higher than pre-infusion levels while glass sphere therapy patients only saw a 7% rise in post-infusion blood levels. The coefficients of variation (COVs) across glass sphere measurements pre, during, and post, were only 0.008 while resin sphere measures saw much greater variability with a COV of 0.45. Both glass and resin therapies showed blood levels at 2-4 h post-injection to be similar to levels measured pre-injection. Neither therapy showed any signs of focal aggregation at 2-4 h post-infusion on whole-body PET/CT. Although glass and resin radioembolization therapies are similar, they both have unique characteristics related to their administration and handling by staff. Understanding the nuances can assist in providing more efficient delivery, better staff education, and reducing radiation exposure to everyone involved with these therapies. The use of near real-time monitoring is feasible and can be used to obtain critical information about the delivery success of a therapy and can inform physicians on their techniques to optimize their practice as well as provide more consistent training to residents.

A Low-Power Radiation Detection SoC With Neural Network Accelerator for Radioisotope Identification

This work presents a low-power radiation detection and radioisotope identification System on Chip (SoC) featuring mixed-signal sensory electronics and on-chip neural network (NN) acceleration hardware. The chip electronics form a multichannel analyzer (MCA) to process the signal from an external radiation detector and produce an energy spectrum of the incoming radiation detections. The NN hardware accelerator then provides a novel means of using the spectrum to perform radioisotope identification. The chip is, thus, able to determine which radioisotopes are present in the radiation source. An on-chip 32-bit microcontroller (MCU) configures the analog front-end (AFE) electronics, oversees the NN hardware, and provides an interface to retrieve the histogram and NN data. The chip is fabricated on a 65-nm CMOS technology and consumes 8.8 mW while acquiring a histogram. In the tested configuration, the radioisotope identification NN takes <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$694~\mu \text{s}$ </tex-math></inline-formula> to execute, consuming <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$1.94~\mu \text{J}$ </tex-math></inline-formula> of energy and using 13.8 kB of RAM in the process. After training, the system correctly identified which isotopes were present in all histograms in a test set composed of combinations of four trained isotopes.

Method for active spatial alignment and stabilization of laser beams in multi-beam systems

We propose a new method for the development of multi-beam systems for the spatial alignment and stability of beams based on the error separation technique. This method avoids alignment errors caused by coupling effect of piezoelectric devices, inaccurate correction calculations, and detection mode of the angular deviation. According to the results by external detectors, the error value of spatial alignment and the root mean square (RMS) of deviations under control during 1 h can be equivalent to approximately 0.87 and 1.06 nm at the sample plane under an oil immersion lens (focal length f=2 mm). The RMS of deviations is less than one-third of those currently reported for multi-beam systems; therefore, higher alignment and stability accuracy can be achieved with our proposed method.

Leveraging Code Snippets to Detect Variations in the Performance of HPC Systems

Variations in the performance of parallel and distributed systems are becoming increasingly challenging. The runtimes of different executions can vary greatly even with a fixed number of computing nodes. Many HPC applications on supercomputers exhibit such variance. This not only leads to unpredictable execution times, but also renders the system’s behavior unintuitive. The efficient online detection of variations in performance is an open problem in HPC research. To solve it, we propose an approach, called <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">vSensor</small> , to detect variations in the performance of systems. The key finding of this study is that the source code of programs can better represent performance at runtime than an external detector. Specifically, many HPC applications contain code snippets that are fixed workload patterns of execution, e.g., the workload of an invariant quantity and a linearly growing workload. This observation allows us to automatically identify these snippets of workload-related code and use them to detect variations in performance. We evaluate <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">vSensor</small> on the Tianhe-2A system with a large number of parallel applications, and the results indicate that it can efficiently identify variations in system performance. The average overhead of 4,096 processes is less than 6% for fixed-workload v-sensors. We identify a problematic node with slow memory by using <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">vSensor</small> that degrades the performance of the program by 21%. A serious issue with network performance is also detected that slows down the Tianhe-2A system by 3.37 times for an HPC kernel.

End2End Occluded Face Recognition by Masking Corrupted Features.

With the recent advancement of deep convolutional neural networks, significant progress has been made in general face recognition. However, the state-of-the-art general face recognition models do not generalize well to occluded face images, which are exactly the common cases in real-world scenarios. The potential reasons are the absences of large-scale occluded face data for training and specific designs for tackling corrupted features brought by occlusions. This article presents a novel face recognition method that is robust to occlusions based on a single end-to-end deep neural network. Our approach, named FROM (Face Recognition with Occlusion Masks), learns to discover the corrupted features from the deep convolutional neural networks, and clean them by the dynamically learned masks. In addition, we construct massive occluded face images to train FROM effectively and efficiently. FROM is simple yet powerful compared to the existing methods that either rely on external detectors to discover the occlusions or employ shallow models which are less discriminative. Experimental results on the LFW, Megaface challenge 1, RMF2, AR dataset and other simulated occluded/masked datasets confirm that FROM dramatically improves the accuracy under occlusions, and generalizes well on general face recognition.