Output Scales Research Articles

Assessing green production efficiency and spatial characteristics of China’s real estate industry based on the undesirable super-SBM model

As China strives to balance rapid urbanization with environmental conservation, increasing attention is being paid to the pursuit of green production efficiency (GPE) in the real estate industry. The undesirable super-SBM model was used to calculate the GPE of China's real estate industry from 2001 to 2020. Additionally, GPE spatial distribution characteristics in China's real estate industry were analyzed using the standard deviation ellipse (SDE), Moran’s index, Theil index, random kernel density estimation (RKDA), and spatial Markov chain (SMC) methods. The GPE exhibited a U-shaped trend, with 2008 as the inflection point, first decreasing and then increasing. It reached a maximum value of 0.747 in 2020. The Theil index increased from 0.043 to 0.121 nationwide, indicating the overall characteristics of low-level slow growth, and imbalance. Discrepancies in input–output scales, the southward shift of economic centers, and population movements contribute significantly to the disparities between the east and west, north and south, and regions divided by the Hu Huanyong Line (Hu Line). The GPE exhibited club convergence characteristics; however, polarization phenomena exist in local areas. Spatial spillover effects were also observed in GPE. Finally, we provide recommendations for promoting green development in the real estate industry, including green building technology, fiscal subsidy investment, and population migration management.

Induced Power Scaling Alone Cannot Explain Griffenfly Gigantism.

Paleozoic skies were ruled by extinct odonatopteran insects called "griffenflies," some with wingspans 3 times that of the largest extant dragonflies and 10 times that of common extant dragonflies. Previous studies suggested that flight was possible for larger fliers because of higher atmospheric oxygen levels, which would have increased air density. We use actuator disk theory to evaluate this hypothesis. Actuator disk theory gives similar estimates of induced power as have been estimated for micro-air vehicles based on insect flight. We calculate that for a given mass of griffenfly, and assuming isometry, a higher density atmosphere would only have reduced the induced power required to hover by 11%, which would have supported a flyer 3% larger in linear dimensions. Steady-level forward flight would have further reduced induced power but could only account for a flier 5% larger in linear dimensions. Further accounting for the higher power available due to high-oxygen air and assuming isometry, we calculate that the largest flyer hovering would have been only 1.19times longer than extant dragonflies. We also consider known allometry in dragonflies and estimated allometry in extinct griffenflies. But such allometry only increases flyer size to 1.22times longer while hovering. We also consider profile and parasite power, but both would have been higher in denser air and thus would not have enhanced the flyability of larger griffenflies. The largest meganeurid griffenflies might have adjusted flight behaviors to reduce power required. Alternatively, the scaling of flight muscle power may have been sufficient to support the power demands of large griffenflies. In literature estimates, mass-specific power output scales as mass0.24 in extant dragonflies. We need only more conservatively assume that mass-specific muscle power scales with mass0, when combined with higher oxygen concentrations and induced power reductions in higher-density air to explain griffenflies 3.4times larger than extant odonates. Experimental measurement of flight muscle power scaling in odonates is necessary to test this hypothesis.

A Risk Assessment Framework Based on Fuzzy Logic for Automotive Systems

Recent advancements in the automotive field have significantly increased the level of complexity and connectivity of modern vehicles. In this context, the topic of cybersecurity becomes extremely relevant, as a successful attack can have an impact in terms of safety on the car navigation, potentially leading to harmful behavior. Risk assessment is typically performed using discrete input and output scales, which can often lead to an identical output in terms of risk evaluation despite the inputs presenting non-negligible differences. This work presents a novel fuzzy-logic-based methodology to assess cybersecurity risks which takes attack feasibility and safety impact as input factors. This technique allows us explicitly model the uncertainty and ambiguousness of input data, which is typical of the risk assessment process, providing an output on a more detailed scale. The fuzzy inference engine is based on a set of control rules expressed in natural language, which is crucial to maintaining the interpretability and traceability of the risk calculation. The proposed framework was applied to a case study extracted from ISO/SAE 21434. The obtained results are in line with the traditional methodology, with the added benefit of also providing the scatter around the calculated value, indicating the risk trend. The proposed method is general and can be applied in the industry independently from the specific case study.

Remote-Sensing Data and Deep-Learning Techniques in Crop Mapping and Yield Prediction: A Systematic Review

Reliable and timely crop-yield prediction and crop mapping are crucial for food security and decision making in the food industry and in agro-environmental management. The global coverage, rich spectral and spatial information and repetitive nature of remote sensing (RS) data have made them effective tools for mapping crop extent and predicting yield before harvesting. Advanced machine-learning methods, particularly deep learning (DL), can accurately represent the complex features essential for crop mapping and yield predictions by accounting for the nonlinear relationships between variables. The DL algorithm has attained remarkable success in different fields of RS and its use in crop monitoring is also increasing. Although a few reviews cover the use of DL techniques in broader RS and agricultural applications, only a small number of references are made to RS-based crop-mapping and yield-prediction studies. A few recently conducted reviews attempted to provide overviews of the applications of DL in crop-yield prediction. However, they did not cover crop mapping and did not consider some of the critical attributes that reveal the essential issues in the field. This study is one of the first in the literature to provide a thorough systematic review of the important scientific works related to state-of-the-art DL techniques and RS in crop mapping and yield estimation. This review systematically identified 90 papers from databases of peer-reviewed scientific publications and comprehensively reviewed the aspects related to the employed platforms, sensors, input features, architectures, frameworks, training data, spatial distributions of study sites, output scales, evaluation metrics and performances. The review suggests that multiple DL-based solutions using different RS data and DL architectures have been developed in recent years, thereby providing reliable solutions for crop mapping and yield prediction. However, challenges related to scarce training data, the development of effective, efficient and generalisable models and the transparency of predictions should be addressed to implement these solutions at scale for diverse locations and crops.

Research on carbon emission factors of heavy chemical industry park based on generalized Divisia index method

Based on the energy consumption and economic data of a heavy chemical industrial park in northwest China from 2013 to 2020, the carbon emissions of the park were calculated. The generalized Divisia index method (GDIM) was employed to analyze the effects of eight absolute and ten relative indicators of carbon emission changes in the heavy chemical industrial park. The results show that optimizing the electricity structure, reducing the energy output and carbon output intensities, and improving investment efficiency have a significant effect on inhibiting carbon emission. The inhibitory effect from reducing investment carbon intensity, fossil energy, and the total energy consumption scale was clearly evident. The deterioration of fossil fuel energy structure, and the increase in investment, output, and total electricity consumption scales are the main factors affecting carbon emission growth. Carbon emissions from industrial processes have a limited impact on carbon emissions. Finally, pertinent suggestions are provided.

New Developments in the High-Energy-Density Solid-Liquid Storage Technology for Redox Flow Batteries

The environmental impact of the use of fossil fuels for energy can be reduced if electricity, which represents one-third of all energy uses, can be generated totally from renewable/sustainable sources such as wind and solar. However, this is only possible if cost-effective long-duration storage technologies are available to allow the highly variable and unpredictable wind and solar energy sources to become reliable baseline energy sources like coal, nuclear or natural gases. Redox flow battery (RFB) energy storage systems are highly suitable for this large-scale, long-duration storage application because while their power output scales with the size of the battery, their energy content resides in the amount of active materials that are stored in external tanks and can be easily scaled up for longer duration.1 The conventional redox flow batteries store electrical energy in the form of some aqueous or non-aqueous soluble ions or compounds in the electrolyte solution. Because of the low solubility (< 2M) of most ions and compounds in aqueous and non-aqueous solvents, these redox flow battery systems have low energy density.2–4 For example, the commercialized all-vanadium RFB system has an average energy density of 20 Wh/kg while that of the lithium-ion battery system is 100-265 Wh/kg.5 To store enough energy for 3-5 days in these RFBs requires a very large volume of solution in a large number of tanks, making these RFB systems expensive due to the cost of tanks and the fluid distribution system and floor space. Our group recently developed a new storage approach that can greatly increase the energy storage density while still enabling the flow battery concept.6 In this approach, the reactants are stored as both soluble ions and their undissolved solid forms and only the liquid containing the soluble ions is circulated through the batteries. This approach potentially enables >4X increase in the storage energy density. This technology was recently demonstrated in a hydrogen-vanadium (VI/V) system, and new test results and findings in this area will be presented in this talk.References H. Zhang, W. Lu, and X. Li, Electrochemical Energy Reviews, 1–15 (2019).D. G. Kwabi et al., Joule, 2, 1894–1906 (2018).M. Wu, T. Zhao, H. Jiang, Y. Zeng, and Y. Ren, Journal of Power Sources, 355, 62–68 (2017).C. Ding, H. Zhang, X. Li, T. Liu, and F. Xing, The Journal of Physical Chemistry Letters, 4, 1281–1294 (2013).A. Manthiram, ACS Central Science, 3, 1063–1069 (2017).Y. Li and T.V. Nguyen, “A Solid-Liquid High-Energy-Density Storage Concept for Redox Flow Batteries and Its Demonstration in an H2-V System,” Paper ID APEN-MIT-2021_023, Applied Energy Symposium: MIT A+B, Aug. 11-13, 2021, Cambridge, MA, USA.

A multi-stage model based on YOLOv3 for defect detection in PV panels based on IR and visible imaging by unmanned aerial vehicle

As solar capacity installed worldwide continues to grow, there is an increasing awareness that advanced inspection systems are becoming of utmost importance to schedule smart interventions and minimize downtime likelihood. In this work we propose a novel automatic multi-stage model to detect panel defects on aerial images captured by unmanned aerial vehicle by using the YOLOv3 network and Computer Vision techniques. The model combines detections of panels and defects to refine its accuracy and exhibits an average inference time per image of 0.98 s. The main novelties are represented by its versatility to process either thermographic or visible images and detect a large variety of defects, to prescript recommended actions to O&M crew to give a more efficient data-driven maintenance strategy and its portability to both rooftop and ground-mounted PV systems and different panel types. The proposed model has been validated on two big PV plants in the south of Italy with an outstanding AP@0.5 exceeding 98% for panel detection, a remarkable AP@0.4 (AP@0.5) of roughly 88.3% (66.9%) for hotspots by means of infrared termography and a mAP@0.5 of almost 70% in the visible spectrum for detection of anomalies including panel shading induced by soiling and bird dropping, delamination, presence of puddles and raised rooftop panels. The model predicts also the severity of hotspot areas based on the estimated temperature gradients, as well as it computes the soiling coverage based on visual images. Finally an analysis of the influence of the different YOLOv3's output scales on the detection is discussed.

An improved YOLO Nano model for dorsal hand vein detection system.

To date, venipuncture, the most necessary and fundamental medical means, still remains a challenging task for medical stuff due to significant individual differences in vein condition. Thanks to mature development in near-infrared (NIR) imaging technology, a series of venepuncture auxiliary equipment has been devised and put into use. Yet, previous researches concentrated more on vein pattern segmentation, failing to materialize the identification of veins suitable to puncture in an embedded system. Given the above, we propose an approach to detect and locate the optimal veins fully utilizing the state-of-the-art deep learning and image processing technologies in order to provide a more practical reference. Firstly, a dedicated NIR-based puncturable vein positioning system is designed, realizing collection of dorsal hand vein images as well as the rapid and accurate location of veins suitable to puncture. Secondly, considering the limitations of embedded devices on computation ability and memory, an improved network based on YOLO Nano, named YOLO Nano-Vein, is presented with architecture trimmed, output scales reduced, and an atrous spatial pyramid pooling (ASPP) added. Finally, average precision (AP) is increased from 91.68 to 93.23%, and the detection time and parameters of network are reduced by 22% and 17.5%, respectively, which validates the proposed network achieves higher accuracy with less detection time in comparison with YOLO Nano and YOLOv3, indicating stronger applicability for detection tasks on embedded devices.

Technologies used by nursing to predict clinical deterioration in hospitalized adults: a scoping review.

to map the early clinical deterioration technologies used in nurses' professional practice in the care of hospitalized adult patients. this is a scoping review, according to Joanna Briggs Institute Reviewer's Manual, which seeks to map the main technologies for detecting early clinical deterioration of hospitalized patients available for use by nurses, summarizing them and indicating gaps in knowledge to be investigated. twenty-seven studies were found. The most present variables in the technologies were vital signs, urinary output, awareness and risk scales, clinical examination and nurses' judgment. The main outcomes were activation of rapid response teams, death, cardiac arrest and admission to critical care units. the study emphasizes the most accurate variables in patient clinical assessment, so that indicative signs of potential severity can be prioritized to guide health conducts aiming to intervene early in the face of ongoing clinical deterioration.

Thermo-Optic Effects in Congruent-LiTaO 3 Based Continuous-Wave Optical Parametric Oscillator

We present an experimental study on single-frequency, singly-resonant optical parametric oscillator (SRO) comprised of 50 mm long congruently-grown, 8-mol%-doped, periodically-poled LiTaO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sub> (MgO:cPPLT) crystal. The SRO is pumped by a single-frequency, fiber laser pumped green laser. Our observations reveal that the SRO output (idler) scales linearly with the pump power ( P <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">p</sub> ) near the SRO threshold ( P <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">p(th)</sub> ). However, the SRO idler power ( P <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i</sub> ) exhibits strong saturation at pump power ≈ 1.5P <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">p(th)</sub> which eventually switches-off the SRO operation on further pump-power enhancement. By coarsely tuning the MgO:cPPLT crystal temperature from 50 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">o</sup> C to 170 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">o</sup> C, the SRO idler wavelength is varied from 1810-1920 nm wavelength band. A maximum idler power P <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i</sub> =583 mW was obtained when the input pump power was ≈ 3.1 W. The idler power saturation is accompanied by discernible distortion in idler beam profile. An early onset of saturation in MgO:cPPLT is essentially a consequence of strong linear absorption exhibited by MgO:cPPLT at pump wavelength, its high thermo-optic coefficient as well as low thermal conductivity. The results are theoretically analyzed by solving the heat conduction equation which provides a deeper insight about the impact of pump-induced thermo-optic effects on a three-wave interaction process.

Fixed cost allocation considering the input-output scale based on DEA approach

As independent and decentralized entities often share common resources, it is necessary to allocate common costs across these subunits. When there is a lack of specific price information, data envelopment analysis (DEA) approaches have been successfully used to resolve fixed cost allocation problems. To this end, this paper develops a fixed cost allocation scheme based on decision-making unit (DMU) input–output scales that considers both the input consumption and output production scales. The allocation scheme is derived from a one-dimensional case to a general case, in which the input–output scale is calculated based on a common set of weights (CSW) determined using a DEA approach. The proposed method is demonstrated using a classical numerical example from the prior literature and an empirical application that has been applied in previous studies, from which it is found that the fixed cost allocation scheme is more acceptable to the involved DMUs.

A Novel High-Energy-Density Storage and Operation Concept for Redox Flow Batteries

The environmental impact of the use of fossil fuels for energy can be greatly reduced if electricity, which represents one-third of all energy uses, can be generated totally from renewable/sustainable sources such as wind and solar. However, this is only possible if cost-effective long-duration (>5 days) storage technologies are available to allow the highly variable and unpredictable wind and solar energy sources to become reliable baseline energy sources like coal or natural gases. Redox flow battery (RFB) energy storage systems are highly suitable for this large-scale, long-duration storage application because while their power output scales with the size of the battery, an expensive component of the storage system, their energy content resides in the amount of active materials that are stored in external tanks and can be easily scaled up for longer duration.1 A stationary battery system like the lithium-ion battery does not have this separation of power and energy characteristic. Therefore, an approach that can significantly increase the operational duration of the RFB systems at a minimal cost is of great interest. The conventional redox flow batteries store electrical energy in the form of some aqueous or non-aqueous soluble ions or compounds in the electrolyte solution. Because of the low solubility (~1M-2M) of most ions and compounds in aqueous and non-aqueous solvents, these redox flow battery systems have low energy density (as compared to solid reactants like lithium in lithium batteries).2–4 For example, the commercialized all-vanadium RFB system has an average energy density of 20 Wh/kg while that of the lithium-ion battery system is 100-265 Wh/kg.5 To store enough energy for 3-5 days in these RFBs requires a very large volume of solution in a large number of tanks, making these RFB systems expensive due to the cost of tanks and the fluid distribution system and floor space. This presentation will discuss a concept that increases the amount of active materials stored in a given storage volume while still enabling the flow battery concept. The novel storage design is to i) store these reactants in a solution of soluble ions and an additional amount of these ions in their undissolved solid form and ii) to circulate only the liquids through the batteries. In their solid form, their energy densities will be much higher than that of the aqueous solution. For example, the concentration of vanadyl sulfate (VOSO4), an active compound used in the all-vanadium RFB system, is around 1.5M. At 1.5M, one has 1.5 moles of vanadium ions per liter or 40.2 Ah/L. In its 6-ligand-water solid form (VOSO4·6H2O) which has a density of 1920 g/L,6 a liter of this solid contains 7.1 moles of vanadium ions or 189 Ah energy. This is a tremendous increase and will put the energy storage density of the RFB systems within those of solid systems like lithium-ion batteries. An existing 6-hr RFB system can become a >24-hr system with minimal modifications. This concept and the processes used to demonstrate it will be discussed in this presentation. A demonstration of this aqueous/solid storage concept in a hydrogen-vanadium flow battery will be presented in another presentation from our group.Acknowledgments This work was also supported by the National Science Foundation under grant number CBET-2024378.

Development and sensibility assessment of a health-related quality of life instrument for adults with severe disabilities who are non-ambulatory.

BackgroundInsight in health‐related quality of life (HRQoL) of adults with severe disabilities who are non‐ambulatory is important, but a measure is lacking. The aim was to develop a HRQoL measure for this group.MethodThe developmental process consisted of the adaptation process of a proxy HRQoL measure for children with severe disabilities who are non‐ambulatory and the assessment of the sensibility of the developed instrument. A three‐step process was used: focus groups, e‐survey and interviews.ResultsIn total, 72% of the items remained unchanged. Three new items and one element to an existing item were added. In ten items, the formulation of the items was adapted to the target group. Concerning the sensibility, respondents suggested minor changes to the instruction and the output scales.ConclusionsThis study has yielded a proxy HRQoL measure for adults with severe disabilities who are non‐ambulatory, the CPADULT, with good sensibility.

Do producers respond to quality information disclosure? The HACCP certification in meat industry

PurposeFood markets are characterized by asymmetric information between suppliers and consumers, which causes inefficiency of market and food safety risks. This paper studies how the food quality and safety information disclosed by the government affects the hazard analysis and critical control point (HACCP) certification decision of meat producers. The heterogeneity of the effects across different regions, provinces with different meat output scales and provinces with different intensities of food safety regulation is evaluated.Design/methodology/approachThis paper applies a unique database comprising information from multiple sources. Food quality and safety information disclosure is indicated by the number of failure records of food sampling inspections by the government in 2015–2018. Fixed-effect model is used in the analyses.FindingsThe results demonstrate that food quality and safety information disclosure has a significant effect on the HACCP certification adoption by meat producers. The effect is heterogeneous across geographic regions, i.e. this effect is larger in the east and the middle of China than that in the west and the northeast. The heterogeneity across regions may be caused by the variance in meat output scales and fiscal expenditures on food safety among provinces.Originality/valueThis research is one of the preliminary attempts to understand how producers respond in terms of HACCP certification to the amount of food quality and safety information disclosed by the government, based on the case of meat industry in China.

Multiplexed electrospray emission on a porous wedge

Linear porous wedge electrospray emitters exhibit a discrete number of emission sites that naturally form during operation. An analytical model is developed to examine the behavior and spacing of these emission sites via the pressure variation in the porous fluid flow associated with the flow focusing on each emission site, which is coupled with the local electric field. The solution for site spacing and current is informed by empirical results with support from electric field modeling and investigation of porous media parameters. Emission site currents of up to 500 nA and site spacings of roughly 50 µm–300 µm are predicted. Results from the model match well with experimental trends and provide further insights into the current and spacing of the discrete emission sites. These insights include the following: (1) for the investigated geometry, the total current can be estimated without taking into account the effects local to each emission site, (2) the wedge hydraulic resistance shows how the emitter output scales with emitter geometry and propellant properties, and (3) the emitted charge to mass ratio increases with the applied electric field. Last, we present a physical description of how specific charge increases with the restorative pressure from the reservoir.

The Tidal Stream Energy Resource of the Fromveur Strait—A Review

Refined assessments of the available tidal stream energy resource are required to optimize turbines design and guarantee successful implementations and operations of devices in the marine environment. Investigations primary focused on identifying areas with maximum current speeds. However, further information may be reached by exhibiting (i) resource temporal variability, (ii) superimposed effects of meteo-oceanographic conditions (including especially wind-generated surface-gravity waves), and (iii) potential environmental impacts of operating turbines at the regional (e.g., changes in sediment transport and surrounding seabed features, effects on marine water quality, etc.) and local (wake-wake interactions and energy output) scales. These aspects are here investigated by reviewing a series of research studies dedicated to the Fromveur Strait off western Brittany, a region with strong potential for tidal array development along the coast of France. Particular attention is dedicated to the exploitation of combined in-situ and remote-sensing observations and numerical simulations. Beyond a site specific characterization of the tidal stream energy resource, this review promotes a series of original approaches and analysis methods for turbines optimization, thus complementing technical specifications to secure the key steps of a tidal energy project and promote the growth of a reliable tidal stream energy exploitation.

2.05 µm chirped pulse amplification system at a 1 kHz repetition rate-2.4 ps pulses with 17 GW peak power.

Ho-doped yttrium lithium fluoride chirped pulse amplification (CPA) is implemented with a high-gain regenerative amplifier (RA) and a two-stage booster amplifier. We demonstrate the generation of 52.5 mJ pulses with a duration of 2.4 ps at a 1 kHz repetition rate. A peak power of 17 GW is achieved for the 2050 nm pulses. The CPA displays a remarkably high stability with a pulse-to-pulse rms as low as 0.23%. The RA operates without any signs of bifurcation and delivers 12 mJ pulses. Seeding the booster amplifier with the RA output scales the pulse energy linearly up into the 50-60 mJ range. The amplifier system is operated at room temperature and shows a high optical-to-optical efficiency of 20.3% with respect to the optical pump power.

New Methods for the Steady-State Analysis of Complex Agent-Based Models

Among all tools used to understand collective human behaviour, few tools have been as successful as agent-based models (ABMs). These models have been particularly effective at describing emergent social behaviour, such as spatial segregation in neighbourhoods or opinion polarization on social networks. ABMs are particularly common in the study of opinion and belief dynamics, being used by fields ranging from anthropology to statistical physics. These models, much like the social systems they describe, often do not have unique output variables, scales, or clear order parameters. This lack of clearly measurable emergent behaviour makes such complex ABMs difficult to study, ultimately limiting their application to cases of empirical interest. In this paper, we introduce a series of approaches to analyze complex multidimensional ABMs, drawing from information theory and cluster analysis. We use these approaches to explore a multi-level agent-based model of ideological alignment introduced by Banisch and Olbrisch to extend Mas and Flache's argument communication theory of bi-polarization. We use the tools introduced here to perform a thorough analysis of the model for small system sizes, identifying the convergence towards steady-state behaviour, and describing the full spectrum of steady-state distributions produced by this model. Finally, we show how the approach we introduced can be easily adapted for larger implementations, as well as for other complex agent-based models of social behaviour.

Technique for inferring angle change as a function of time for high-current electron beams using a dose-rate monitor array.

Intense electron beams striking a high-atomic number target produce high-output pulsed photon fluxes for flash x-ray experiments. Without an external guide field, such beams are subject to the dynamics of high-current electron beam propagation, including changes to electron trajectories either from self-fields or from development of beam instabilities. The bremsstrahlung output (dose-rate) scales approximately as IVx, where I is the beam current, V the electron energy, and x is in the range 2.0-2.65 and depends upon the electron angle on the converter. Using experimental beam data (dose-rate, I and V), this equation can be solved for x, a process known as "inverting the radiographer's equation." Inversion methods that rely on thermoluminescent dosimeters, which are time-integrated, yield no information about evolution of the electron beam angle in time. We propose here an inversion method that uses several dose-rate monitors at different angles with respect to the beam axis. By measuring dose-rates at different angles, one can infer the time-dependent beam voltage and angle. This method compares well with estimates of corrected voltage and results in a self-consistent picture of beam dynamics. Techniques are demonstrated using data from self-magnetic pinch experiments at the RITS-6 facility at Sandia National Laboratories.

On the benefit of adversarial training for monocular depth estimation

In this paper we address the benefit of adding adversarial training to the task of monocular depth estimation. A model can be trained in a self-supervised setting on stereo pairs of images, where depth (disparities) are an intermediate result in a right-to-left image reconstruction pipeline. For the quality of the image reconstruction and disparity prediction, a combination of different losses is used, including L1 image reconstruction losses and left–right disparity smoothness. These are local pixel-wise losses, while depth prediction requires global consistency. Therefore, we extend the self-supervised network to become a Generative Adversarial Network (GAN), by including a discriminator which should tell apart reconstructed (fake) images from real images. We evaluate Vanilla GANs, LSGANs and Wasserstein GANs in combination with different pixel-wise reconstruction losses. Based on extensive experimental evaluation, we conclude that adversarial training is beneficial if and only if the reconstruction loss is not too constrained. Even though adversarial training seems promising because it promotes global consistency, non-adversarial training outperforms (or is on par with) any method trained with a GAN when a constrained reconstruction loss is used in combination with batch normalisation. Based on the insights of our experimental evaluation we obtain state-of-the art monocular depth estimation results by using batch normalisation and different output scales.