Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “ARTIFACTS”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 217 records · Page 12

Probing for Artifacts: Detecting Imagenet Model Evasions

While deep learning models have made incredible progress across a variety of machine learning tasks, they remain vulnerable to adversarial examples crafted to fool otherwise trustworthy models. In this work we approach this problem through the lens of a detection framework. We propose a classification network that uses the hidden layer activations of a trained model as inputs to detect adversarial artifacts in an input. We train this classification network simultaneously against multiple adversarial algorithms to create a more robust detector and show higher detection rates than several alternatives. The novelty of our approach is in the scale and scope of probing Imagenet models for adversarial artifacts. In addition, we propose an improvement to feature squeezing, another common adversarial example detection method.

Rounds, Jeremiah↗

Qualitative and quantitative effects of harmonic echocardiographic imaging on endocardial edge definition and side-lobe artifacts

Harmonic imaging is a new ultrasonographic technique that is designed to improve image quality by exploiting the spontaneous generation of higher frequencies as ultrasound propagates through tissue. We studied 51 difficult-to-image patients with blinded side-by-side cineloop evaluation of endocardial border definition by harmonic versus fundamental imaging. In addition, quantitative intensities from cavity versus wall were compared for harmonic versus fundamental imaging. Harmonic imaging improved left ventricular endocardial border delineation over fundamental imaging (superior: harmonic = 71.1%, fundamental = 18.7%; similar: 10.2%; P <.001). Quantitative analysis of 100 wall/cavity combinations demonstrated brighter wall segments and more strikingly darker cavities during harmonic imaging (cavity intensity on a 0 to 255 scale: fundamental = 15.6 +/- 8.6; harmonic = 6.0 +/- 5.3; P <.0001), which led to enhanced contrast between the wall and cavity (1.89 versus 1.19, P <.0001). Harmonic imaging reduces side-lobe artifacts, resulting in a darker cavity and brighter walls, thereby improving image contrast and endocardial delineation.

NASA Discipline Cardiopulmonary↗

Synchrotron radiation-based x-ray analysis of bronze artifacts from an Iron Age site in the Judean hills.

Seven bronze bangles from Tell en-Nasbeh, northern Judah, were investigated to understand the phase composition and manufacturing process of the artifacts, and possibly suggest a provenance for their origin. Synchrotron x-ray radiation diffraction (XRD) and fluorescence (XRF) were used in the analysis to avoid any destructive sampling and at the same time penetrate through the surface into the core metal. These techniques enabled us to determine that the bangles were not just tin bronze, but leaded tin bronze. Based on excavation reports, it is unlikely that the metal objects were manufactured locally at Tell en-Nasbeh; rather, preliminary XRD and XRF data point towards the neighboring region of Edom as their origin. Despite their political enmity during the Iron Age II, the data suggest that Judahite social demands for bronze may have fostered a strong economic relationship between these two polities.

43 PARTICLE ACCELERATORS↗

A new oxidation based technique for artifact free TEM specimen preparation of nuclear graphite

Graphite is a key component in designs of current and future nuclear reactors whose in-service lifetimes are dependent upon the mechanical performance of the graphite. Irradiation damage from fast neutrons creates lattice defects which have a dynamic effect on the microstructure and mechanical properties of graphite. Transmission electron microscopy (TEM) can offer real-time monitoring of the dynamic atomic-level response of graphite subjected to irradiation; however, conventional TEM specimen-preparation techniques, such as argon ion milling itself, damage the graphite specimen and introduce lattice defects. It is impossible to distinguish these defects from the ones created by electron or neutron irradiation. Thus,to ensure that TEM specimens are artifact-free, a new oxidation-based technique has been developed. Bulk nuclear grades of graphite (IG-110 and NBG-18) and highly oriented pyrolytic graphite (HOPG) were initially mechanically thinned to ~60μm. Discs 3mm in diameter were then oxidized at temperatures between 575°C and 625°C in oxidizing gasses using a new jet-polisher-like set-up in order to achieve optimal oxidation conditions to create self-supporting electron-transparent TEM specimens. The quality of these oxidized specimens were established using optical and electron microscopy. Samples oxidized at 575°C exhibited large areas of electron transparency and the corresponding lattice imaging showed no apparent damage to the graphite lattice.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Origin and Suppression of Beam Damage-Induced Oxygen-K Edge Artifact from γ-Al 2 O 3 using Cryo-EELS

Gamma-alumina (γ-Al 2 O 3 ), like other low-Z oxides, is readily damaged when exposed to an electron beam. This typically results in the formation of a characteristic pre-edge peak in the oxygen-K edge of electron energy-loss spectra (EELS) acquired during or after the damage process. This artifact can mask the presence of intrinsic O-K edge fine structure that would reveal chemical properties of the material; therefore, its suppression is key. In this work, we systematically investigate the conditions that give rise to the damage-induced O-K pre-edge peak and show that it can be effectively suppressed by performing EELS experiments at cryogenic (cryo) temperatures. Prolonged exposure of γ-Al 2 O 3 to a focused electron beam results in a hole bored through the sample; this was used as a reproducible beam damage condition. O-K edge EELS spectra were collected from a single-crystal γ-Al 2 O 3 sample both during and after focused 2 electron beam hole drilling, and at room and cryo temperatures, using a monochromated scanning transmission electron microscope (STEM). Furthermore, the characteristic 531 eV pre-edge peak visible in the room temperature EELS spectra was completely suppressed in the cryo-EELS spectra, even in the presence of a visible drilled hole. We then correlated these experimental observations with multiple-scattering EELS simulations to determine the likely atomistic origin of the damage-induced O-K pre-edge peak. The findings indicate that the pre-edge peak is caused primarily by the presence of surface O dimer (O-O) bonds formed during beam damage, and that operating at cryo temperature suppresses the formation of surface O-O bonds, thus preventing formation of the O-K pre-edge peak. Additionally, Al-L 2,3 edge EELS spectra revealed Al loss primarily from tetrahedral sites during hole drilling.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

Characterization of Sulfur and Nanostructured Sulfur Battery Cathodes in Electron Microscopy Without Sublimation Artifacts

Abstract Lithium sulfur (Li–S) batteries have the potential to provide higher energy storage density at lower cost than conventional lithium ion batteries. A key challenge for Li–S batteries is the loss of sulfur to the electrolyte during cycling. This loss can be mitigated by sequestering the sulfur in nanostructured carbon–sulfur composites. The nanoscale characterization of the sulfur distribution within these complex nanostructured electrodes is normally performed by electron microscopy, but sulfur sublimates and redistributes in the high-vacuum conditions of conventional electron microscopes. The resulting sublimation artifacts render characterization of sulfur in conventional electron microscopes problematic and unreliable. Here, we demonstrate two techniques, cryogenic transmission electron microscopy (cryo-TEM) and scanning electron microscopy in air (airSEM), that enable the reliable characterization of sulfur across multiple length scales by suppressing sulfur sublimation. We use cryo-TEM and airSEM to examine carbon–sulfur composites synthesized for use as Li–S battery cathodes, noting several cases where the commonly employed sulfur melt infusion method is highly inefficient at infiltrating sulfur into porous carbon hosts.

25 ENERGY STORAGE↗

The Application and Potential Artifacts of Zeeman Cold Vapor Atomic Absorption Spectrometry in Mercury Stable Isotope Analysis

Zeeman cold vapor atomic absorption spectrometry (CVAAS) has been widely used for environmental mercury (Hg) detection and quantification for decades, but little is known about its utility and potential artifacts in analyzing Hg with varying isotope compositions. We show that each Hg isotope responds differently by CVAAS analysis, with 200 Hg and 202 Hg isotopes exhibiting signal intensities ~10 times greater than those of 198 Hg and 201 Hg isotopes. However, all Hg isotopes show a linear correlation between Hg concentration and signal intensity, validated by both measurements and theoretical simulations. Zeeman CVAAS could thus offer a convenient, inexpensive tool for determining Hg isotopes, particularly in using one or two enriched Hg isotopes for tracing Hg biogeochemical transformations, such as partitioning, ion exchange, sorption–desorption, and methylation–demethylation in environmental matrices. Additionally, we caution that care must be taken when CVAAS is used for quantifying Hg in samples with changing isotope compositions to avoid measurement errors.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Reduced Artifact Approach for Determining Diffusion Coefficients in Time-Resolved Microscopy

Ultrafast microscopy methods traditionally assume a Gaussian profile to extract excited state diffusivities from transport measurements. Although this fitting method recovers accurate diffusion coefficients when the point spread function is well-represented by a Gaussian, even minor spatial aberrations introduced by the imaging system cause significant errors in the determined value. To provide a more accurate measure of excited state transport in nano- and microscale materials systems, in this work an alternative analysis protocol is proposed that numerically convolves the Green’s function solution to the diffusion equation with the experimentally measured point spread function. In contrast to the Gaussian fitting approach, the numerical convolution is shown to be robust against artifacts caused by nonideal point spread functions. Furthermore, the numerical convolution approach is highly effective at resolving anisotropic diffusion in modeled data.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

MRI Artifacts of a Metallic Stent Derived From a Human Aorta Specimen

Magnetic resonance imaging has proved to be a useful technique to get images of the whole body. However, the presence of ferromagnetic material can cause susceptibility artifacts, which result from microscopic gradients that occur near the boundaries between areas displaying different magnetic susceptibility. These gradients cause dephasing of spins and frequency shifts in the surrounding tissues. Intravoxel dephasing and spatial mis-registration can degrade image quality. An aorta with a metallic stent was preserved in formaldehyde at 10% inside acrylic cylinders and used to obtain MR images. We tested pulsed spin echo and gradient echo sequences to improve image quality. All experiments were performed on a 7T/21 cm Varian system (Varian, Inc, Palo Alto, CA) equipped with Direct Drive technology and a 16-rung birdcage coil transceiver. The presence of metallic stents produces a lack of signal that might give falsely reassuring appearances within the vessel lumen.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Artifacts in X-ray Dark-Field Tomography

Grating-based x-ray imaging provides three principle kinds of contrast: absorption, phase, and dark-field. Due to the availability of tomographic reconstruction algorithms for the dark-field contrast, it is now possible to take advantage of quantitative scatter information. However, the published algorithm is based on several assumptions that might be violated in reality. We use numerical simulations in order to identify artifacts in the reconstructions, which is crucial for the interpretation of experimental data.

07 ISOTOPE AND RADIATION SOURCES↗

Completely inverted hysteresis loops: Inhomogeneity effects or experimental artifacts

Completely inverted hysteresis loops (IHL) are obtained by the superconducting quantum interference device with large cooling fields (>10 kOe) in (La,Sr)MnO{sub 3} films with self-assembled LaSrMnO{sub 4}, an antiferromagnetic interface. Although the behaviours of measured loops show many features characteristic to the IHL, its origin, however, is not due to the exchange coupling between (La,Sr)MnO{sub 3}/LaSrMnO{sub 4}, an often accepted view on IHL. Instead, we demonstrate that the negative remanence arises from the hysteresis of superconducting coils, which drops abruptly when lower cooling fields are utilized. Hence the completely inverted hysteresis loops are experimental artifacts rather than previously proposed inhomogeneity effects in complicated materials.

36 MATERIALS SCIENCE↗

Artifact mitigation of ptychography integrated with on-the-fly scanning probe microscopy

In this paper, we report our experiences with conducting ptychography simultaneously with the X-ray fluorescence measurement using the on-the-fly mode for efficient multi-modality imaging. We demonstrate that the periodic artifact inherent to the raster scan pattern can be mitigated using a sufficiently fine scan step size to provide an overlap ratio of >70%. This allows us to obtain transmitted phase contrast images with enhanced spatial resolution from ptychography while maintaining the fluorescence imaging with continuous-motion scans on pixelated grids. Lastly, this capability will greatly improve the competence and throughput of scanning probe X-ray microscopy.

36 MATERIALS SCIENCE↗

Wake Capture, Particle Breakup, and Other Artifacts Associated with Counterflow Virtual Impaction

Counterflow virtual impaction is used to inertially separate cloud elements from un-activated aerosol. Previous airborne, ground-based and laboratory studies using this technique exhibit artifacts that are not fully explained by impaction theory. We have performed laboratory studies that show small particles can be carried across the inertial barrier of the counterflow by collision and/or coalescence or riding the wake of larger particles with sufficient inertia. We have also performed theoretical calculations to show that aerodynamic forces associated with the requisite acceleration and deceleration of particles within a counterflow virtual impactor can lead to breakup. The implication of these processes on studies using this technique are discussed.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Inversion of Many-Beam Bragg Intensities for Phasing by Iterated Projections: Removal of Multiple Scattering Artifacts from Diffraction Data

An iterated projection algorithm (N-Phaser) is developed that reconstructs a scattering potential from N-beam multiple Bragg scattered intensities. The method may be used to eliminate multiple scattering artifacts from electron diffraction data, solving the phase problem and increasing the thicknesses of samples used in materials science, solid-state chemistry, and small molecule crystallography. For high-energy transmission electron diffraction, we show that the algorithm recovers accurate complex structure factors from a wide range of thicknesses, orientations, and relativistic beam energies, and does not require known thickness or atomic-resolution data if sufficient multiple scattering occurs. Furthermore, extensions to Cryo-electron microscopy and Micro-electron diffraction are suggested.

36 MATERIALS SCIENCE↗

Artifact identification in X-ray diffraction data using machine learning methods

In situ synchrotron high-energy X-ray powder diffraction (XRD) is highly utilized by researchers to analyze the crystallographic structures of materials in functional devices ( e.g. battery materials) or in complex sample environments ( e.g. diamond anvil cells or syntheses reactors). An atomic structure of a material can be identified by its diffraction pattern along with a detailed analysis of the Rietveld refinement which yields rich information on the structure and the material, such as crystallite size, microstrain and defects. For in situ experiments, a series of XRD images is usually collected on the same sample under different conditions ( e.g. adiabatic conditions) yielding different states of matter, or is simply collected continuously as a function of time to track the change of a sample during a chemical or physical process. In situ experiments are usually performed with area detectors and collect images composed of diffraction patterns. For an ideal powder, the diffraction pattern should be a series of concentric Debye–Scherrer rings with evenly distributed intensities in each ring. For a realistic sample, one may observe different characteristics other than the typical ring pattern, such as textures or preferred orientations and single-crystal diffraction spots. Textures or preferred orientations usually have several parts of a ring that are more intense than the rest, whereas single-crystal diffraction spots are localized intense spots owing to diffraction of large crystals, typically >10 µm. In this work, an investigation of machine learning methods is presented for fast and reliable identification and separation of the single-crystal diffraction spots in XRD images. The exclusion of artifacts during an XRD image integration process allows a precise analysis of the powder diffraction rings of interest. When it is trained with small subsets of highly diverse datasets, the gradient boosting method can consistently produce high-accuracy results. The method dramatically decreases the amount of time spent identifying and separating single-crystal diffraction spots in comparison with the conventional method.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗