Results for “Pixel of Interest” — Page 10

NASA NTRS2011

Hurricane Data Analysis Tool

In order to facilitate Earth science data access, the NASA Goddard Earth Sciences Data Information Services Center (GES DISC) has developed a web prototype, the Hurricane Data Analysis Tool (HDAT; URL: http://disc.gsfc.nasa.gov/HDAT), to allow users to conduct online visualization and analysis of several remote sensing and model datasets for educational activities and studies of tropical cyclones and other weather phenomena. With a web browser and few mouse clicks, users can have a full access to terabytes of data and generate 2-D or time-series plots and animation without downloading any software and data. HDAT includes data from the NASA Tropical Rainfall Measuring Mission (TRMM), the NASA Quick Scatterometer(QuikSCAT) and NECP Reanalysis, and the NCEP/CPC half-hourly, 4-km Global (60 N - 60 S) IR Dataset. The GES DISC archives TRMM data. The daily global rainfall product derived from the 3-hourly multi-satellite precipitation product (3B42 V6) is available in HDAT. The TRMM Microwave Imager (TMI) sea surface temperature from the Remote Sensing Systems is in HDAT as well. The NASA QuikSCAT ocean surface wind and the NCEP Reanalysis provide ocean surface and atmospheric conditions, respectively. The global merged IR product, also known as, the NCEP/CPC half-hourly, 4-km Global (60 N -60 S) IR Dataset, is one of TRMM ancillary datasets. They are globally-merged pixel-resolution IR brightness temperature data (equivalent blackbody temperatures), merged from all available geostationary satellites (GOES-8/10, METEOSAT-7/5 & GMS). The GES DISC has collected over 10 years of the data beginning from February of 2000. This high temporal resolution (every 30 minutes) dataset not only provides additional background information to TRMM and other satellite missions, but also allows observing a wide range of meteorological phenomena from space, such as, hurricanes, typhoons, tropical cyclones, mesoscale convection system, etc. Basic functions include selection of area of interest and time, single imagery, overlay of two different products, animation,a time skip capability and different image size outputs. Users can save an animation as a file (animated gif) and import it in other presentation software, such as, Microsoft PowerPoint. Since the tool can directly access the real data, more features and functionality can be added in the future.

Liu, Zhong↗

NASA NTRS2004

Event-Driven Random-Access-Windowing CCD Imaging System

A charge-coupled-device (CCD) based high-speed imaging system, called a realtime, event-driven (RARE) camera, is undergoing development. This camera is capable of readout from multiple subwindows [also known as regions of interest (ROIs)] within the CCD field of view. Both the sizes and the locations of the ROIs can be controlled in real time and can be changed at the camera frame rate. The predecessor of this camera was described in High-Frame-Rate CCD Camera Having Subwindow Capability (NPO- 30564) NASA Tech Briefs, Vol. 26, No. 12 (December 2002), page 26. The architecture of the prior camera requires tight coupling between camera control logic and an external host computer that provides commands for camera operation and processes pixels from the camera. This tight coupling limits the attainable frame rate and functionality of the camera. The design of the present camera loosens this coupling to increase the achievable frame rate and functionality. From a host computer perspective, the readout operation in the prior camera was defined on a per-line basis; in this camera, it is defined on a per-ROI basis. In addition, the camera includes internal timing circuitry. This combination of features enables real-time, event-driven operation for adaptive control of the camera. Hence, this camera is well suited for applications requiring autonomous control of multiple ROIs to track multiple targets moving throughout the CCD field of view. Additionally, by eliminating the need for control intervention by the host computer during the pixel readout, the present design reduces ROI-readout times to attain higher frame rates. This camera (see figure) includes an imager card consisting of a commercial CCD imager and two signal-processor chips. The imager card converts transistor/ transistor-logic (TTL)-level signals from a field programmable gate array (FPGA) controller card. These signals are transmitted to the imager card via a low-voltage differential signaling (LVDS) cable assembly. The FPGA controller card is connected to the host computer via a standard peripheral component interface (PCI).

Monacos, Steve↗

NASA NTRS1981

Precision in the evaluation of Landsat autocorrelation - The terrain effect

The autocorrelation present in Landsat data can be described by a two-parameter model. Presence of this autocorrelation seriously inflates estimates of the variance of a set of pixels. The degree of inflation is always serious but varies markedly according to the values of the parameters phi and Theta. Thus corrections of the effects of the model require precise estimates of these parameters. Two hypotheses are proposed to explain the variation of phi with location. Several lines of evidence are presented which support the idea that phi is induced directly by the autocorrelation structure of the terrain being sensed. It is suggested that use of this relation will allow economical estimates of phi for any scene of interest.

Craig, R. G.↗

NASA NTRS2018

Imaging for Hypersonic Experimental Aeroheating Testing (IHEAT) Version 4.0: User Manual

The IHEAT v4.0 software is a data reduction code for global thermography data acquired in the NASA Langley Aerothermodynamics Laboratory (LAL) hypersonic wind tunnels. IHEAT uses red and green color-intensity data from two-dimensional images of wind tunnel models to compute temperatures and heat-transfer rates using a semi-infinite, one-dimensional heat transfer approximation at each image pixel. Multiple automated tools in IHEAT v4.0 decrease the time required to reduce the data from a phosphor thermography wind tunnel run. Data at one or all of the image pixel locations can be exported to computer files for further analysis. The prior version of IHEAT, v3.2, was written in PV-WAVE® (now owned by Rogue Wave® Software) in 1994 and was limited in functionality to fit within the memory constraints of the available computers at the time. IHEAT v4.0 is written in MATLAB® by MathWorks® and contains several new features that leverage the increase in available memory of the current computers. A Piecewise tool permits the user to extract data along a segmented line cut that can follow interesting features in the image better than the single, straight line cuts that were possible with the legacy Length and Profile tools. The new Load Run and Batch tools facilitate batch processing by loading in all of the input files and images for a run at the same time. Load Run permits the user to process the available run images manually, while Batch automatically saves heat transfer data from all of the images based on the analysis previously performed on a single frame. IHEAT v4.0 also can automatically calculate the temporal collapse of reference line cuts from the time history heating data for a run to indicate the appropriate frame to reduce for each run. The IHEAT v4.0 source code was compiled into a standalone executable file that can be accessed remotely from several computers with different operating systems, simultaneously. The software is run through the MATLAB® Compiler Runtime engine, and therefore, IHEAT does not require a software license to run. Any software commands executed in the IHEAT v4.0 code will not affect other similar applications running on the same machine. Similarly, changes to the parent software do not affect a compiled code. These features of IHEAT v4.0 are improvements over the legacy v3.2 code, which required regular maintenance to avoid losing functionality as the PVWAVE ® programming language was upgraded.

Mason, Michelle L.↗

NASA NTRS2011

Enhancements in Deriving Smoke Emission Coefficients from Fire Radiative Power Measurements

Smoke emissions have long been quantified after-the-fact by simple multiplication of burned area, biomass density, fraction of above-ground biomass, and burn efficiency. A new algorithm has been suggested, as described in Ichoku & Kaufman (2005), for use in calculating smoke emissions directly from fire radiative power (FRP) measurements such that the latency and uncertainty associated with the previously listed variables are avoided. Application of this new, simpler and more direct algorithm is automatic, based only on a fire's FRP measurement and a predetermined coefficient of smoke emission for a given location. Attaining accurate coefficients of smoke emission is therefore critical to the success of this algorithm. In the aforementioned paper, an initial effort was made to derive coefficients of smoke emission for different large regions of interest using calculations of smoke emission rates from MODIS FRP and aerosol optical depth (AOD) measurements. Further work had resulted in a first draft of a 1 1 resolution map of these coefficients. This poster will present the work done to refine this algorithm toward the first production of global smoke emission coefficients. Main updates in the algorithm include: 1) inclusion of wind vectors to help refine several parameters, 2) defining new methods for calculating the fire-emitted AOD fractions, and 3) calculating smoke emission rates on a per-pixel basis and aggregating to grid cells instead of doing so later on in the process. In addition to a presentation of the methodology used to derive this product, maps displaying preliminary results as well as an outline of the future application of such a product into specific research opportunities will be shown.

Ellison, Luke↗

NASA NTRS1994

Ice surface temperature retrieval from AVHRR, ATSR, and passive microwave satellite data: Algorithm development and application

One essential parameter used in the estimation of radiative and turbulent heat fluxes from satellite data is surface temperature. Sea and land surface temperature (SST and LST) retrieval algorithms that utilize the thermal infrared portion of the spectrum have been developed, with the degree of success dependent primarily upon the variability of the surface and atmospheric characteristics. However, little effort has been directed to the retrieval of the sea ice surface temperature (IST) in the Arctic and Antarctic pack ice or the ice sheet surface temperature over Antarctica and Greenland. The reason is not one of methodology, but rather our limited knowledge of atmospheric temperature, humidity, and aerosol vertical, spatial and temporal distributions, the microphysical properties of polar clouds, and the spectral characteristics of snow, ice, and water surfaces. Over the open ocean the surface is warm, dark, and relatively homogeneous. This makes SST retrieval, including cloud clearing, a fairly straightforward task. Over the ice, however, the surface within a single satellite pixel is likely to be highly heterogeneous, a mixture of ice of various thicknesses, open water, and snow cover in the case of sea ice. Additionally, the Arctic is cloudy - very cloudy - with typical cloud cover amounts ranging from 60-90 percent. There are few observations of cloud cover amounts over Antarctica. The goal of this research is to increase our knowledge of surface temperature patterns and magnitudes in both polar regions, by examining existing data and improving our ability to use satellite data as a monitoring tool. Four instruments are of interest in this study: the AVHRR, ATSR, SMMR, and SSM/I. Our objectives are as follows. Refine the existing AVHRR retrieval algorithm defined in Key and Haefliger (1992; hereafter KH92) and applied elsewhere. Develop a method for IST retrieval from ATSR data similar to the one used for SST. Further investigate the possibility of estimating surface temperature from passive microwave data (in conjunction with AVHRR clear sky samples) through the use of 'effective emissivities' and physical relationships between skin temperature and subsurface temperature. Use the general method outlined in MK93 to calculate a 12-year record of clear sky equivalent surface temperatures, or possibly all-sky snow-ice interface physical temperatures, from SMMR and SSM/I, compare these temperatures to climatologies, ECMWF modeled surface temperatures, and surface temperatures predicted by a 2-D ice model. And intercompare several ice surface retrieval methods and validate them against ground measurements from the Swiss Camp on the Greenland ice sheet. Additionally, we intend to develop a surface temperature product based on AVHRR data and possibly blended with drifting buoy and meteorological station temperatures.

Key, Jeff↗

DOE OSTI2024

Optical image analysis for graphene layer detection: Enhanced green channel methodology

Graphene, a material of increasing research interest, requires accurate layer identification due to its sensitivity to layer count. Existing methods for graphene layer number identification are either time-consuming or of low accuracy, with high-accuracy methods often requiring expensive processes. This paper aims to address this challenge by proposing a cost-effective and efficient approach. Specifically, the current work highlights only the green channel—one of the three primary color channels (red, green, blue) that make up an optical image—from images of exfoliated graphene flakes for layer count identification. A linear regression is performed between pixel position and substrate green channel value, and this effect is subtracted from the entire optical image to mitigate background effects. By storing the range of green channel values for each type of flake (monolayer, bilayer, or tri-layer) based on a few images, we establish thresholds for identifying different types of layers in a particular setup. Additionally, our methodology allows for flexible threshold tuning using a single reference image, enabling adjustment to changes in detection setup such as illumination level, magnification, or microscope used. Finally, demonstrating high accuracy and flexibility, this methodology presents a suitable technique for graphene layer number identification without the need for large datasets or expensive instruments.

2D materials↗

DOE OSTI2023

End-to-End Pipeline for Trigger Detection on Hit and Track Graphs

There has been a surge of interest in applying deep learning in particle and nuclear physics to replace labor-intensive offline data analysis with automated online machine learning tasks. This paper details a novel AI-enabled triggering solution for physics experiments in Relativistic Heavy Ion Collider and future Electron-Ion Collider. The triggering system consists of a comprehensive end-to-end pipeline based on Graph Neural Networks that classifies trigger events versus background events, makes online decisions to retain signal data, and enables efficient data acquisition. Here, the triggering system first starts with the coordinates of pixel hits lit up by passing particles in the detector, applies three stages of event processing (hits clustering, track reconstruction, and trigger detection), and labels all processed events with the binary tag of trigger versus background events. By switching among different objective functions, we train the Graph Neural Networks in the pipeline to solve multiple tasks: the edge-level track reconstruction problem, the edge-level track adjacency matrix prediction, and the graph-level trigger detection problem. We propose a novel method to treat the events as track-graphs instead of hit-graphs. This method focuses on intertrack relations and is driven by underlying physics processing. As a result, it attains a solid performance (around 72% accuracy) for trigger detection and outperforms the baseline method using hit-graphs by 2% higher accuracy.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS2005

Effect of Forest Canopy on Remote Sensing Soil Moisture at L-band

Global maps of soil moisture are needed to improve understanding and prediction of the global water and energy cycles. Accuracy requirements imply the use of lower frequencies (L-band) to achieve adequate penetration into the soil and to minimize attenuation by the vegetation canopy and effects of surface roughness. Success has been demonstrated over agricultural areas, but canopies with high biomass (e.g. forests) still present a challenge. Examples from recent measurements over forests with the L-band radiometer, 2D-STAR, and its predecessor, ESTAR, will be presented to illustrate the problem. ESTAR and 2D-STAR are aircraft-based synthetic aperture radiometers developed to help resolve both the engineering and algorithm issues associated with future remote sensing of soil moisture. ESTAR, which does imaging across track, was developed to demonstrate the viability of aperture synthesis for remote sensing. The instrument has participated several soil moisture experiments (e.g. at the Little Washita Watershed in 1992 and the Southern Great Plains experiments in 1997 and 1999). In addition, measurements have been made at a forest site near Waverly, VA which contains conifer forests with a variety of biomass. These data have demonstrated the success of retrieving soil moisture at L-band over agricultural areas and the response of passive observations at L-band to biomass over forests. 2D-STAR is a second generation instrument that does aperture synthesis in two dimensions (along track and cross track) and is dual polarized. This instrument has the potential to provide measurements at L-band that simulate the measurements that will be made by the two L-band sensors currently being developed for future remote sensing of soil moisture from space: Hydros (conical scan and real aperture) and SMOS (multiple incidence angle and synthetic aperture). 2D-STAR participated in the SMEX-03 soil moisture experiment, providing images from the NASA P-3 aircraft. Preliminary results include images of the experiment site area near Huntsville, AL that included a mixture of forest and agriculture. Changes during a rain event further illustrate the issues presented by forests. Work is continuing to reduce the 2D-STAR data and to support the two future remote sensing missions. Among the goals is to process the 2D-STAR data to create multiple looks (at the same pixel) with different incidence angles. Data in this format can be used to test algorithms for retrieving soil moisture and biomass such as are planned for SMOS. Also, the data are being processed to provide images at constant incidence angles such as will be obtained by Hydros. Although Hydros will have only one incidence angle, it will also carry an L-band radar, The goal is to use the radar to improve spatial resolution, an issue for remote sensing from space at the long wavelengths. Simultaneous observations with active and passive sensors also offers interesting prospects for treating areas of high biomass (forests) and irregular terrain and may be the challenge for the future.

LeVine, D. M.↗

DOE OSTI2024

Understanding Formation of Irradiation-Induced Defects through 4D-STEM, Electron Tomography, and WBDF-STEM

A major challenge in advancing nuclear materials for next-generation fission and proposed fusion reactors is to comprehensively understand the formation of irradiation-induced defects. Here it is essential to correlate the evolution of irradiation-induced defects and the degradation of mechanical properties, as they collectively dictate the material's lifespan and ensure nuclear safety. Scanning transmission electron microscopy (STEM) based techniques have emerged as indispensable tools for irradiation-induced defect characterization, offering high spatial resolution imaging and chemical analysis, such as electron energy loss spectroscopy (EELS) and energy dispersive X-ray spectroscopy (EDXS). These techniques have been effectively used to obtain an atomic-scale view of the defect structure. Recent advances in electron microscopy, particularly in 4D-STEM, offer detailed insight into microstructural evolution by capturing full 2D diffraction patterns at every pixel position. Using high-speed direct electron detectors, this technology generates a four-dimensional dataset, overcoming the limitations of traditional STEM imaging.

36 MATERIALS SCIENCE↗

NASA NTRS1985

Strategies for analyzing mixed pixels in remotely sensed imagery

Two uncalibrated synthetic aperture radar (SAR) scenes from Death Valley, California, and two uncalibrated scenes from Owens Valley, California, were used to test a technique to identify and separate the primary factors associated with multipolarized radar image sets. Unique to the technique is the concept that varied types of radar polarization signatures are a result of the interaction of a few physical factors (e.g., roughness, volume scattering, and dielectric constant). Thus, the varied signatures observed in an image reflect the interaction between these factors. The objectives are to first separate the interaction of factors directly influencing the measurement and to attempt to identify these factors. Interpretation of the SAR image sets proceeds by testing models that hypothesize the cause and effect relations between those factors directly affecting the radar measurement and those features of interest to the interpreter. The results of our analysis indicated two consistent endmember types over all four image scenes. These types are defined by a high flat polarized signature near a level of 255 DN and a low level flat signature near 0 DN for all polarizations. Four other endmember types were also isolated. The spatial distribution of these endmember types indicates possible correspondence to surface volume scattering and changes in surface dielectric. However, the shapes of these endmember types were different from scene to scene.

Smith, M.↗

NASA NTRS1991

Target Detection Using Fractal Geometry

The concepts and theory of fractal geometry were applied to the problem of segmenting a 256 x 256 pixel image so that manmade objects could be extracted from natural backgrounds. The two most important measurements necessary to extract these manmade objects were fractal dimension and lacunarity. Provision was made to pass the manmade portion to a lookup table for subsequent identification. A computer program was written to construct cloud backgrounds of fractal dimensions which were allowed to vary between 2.2 and 2.8. Images of three model space targets were combined with these backgrounds to provide a data set for testing the validity of the approach. Once the data set was constructed, computer programs were written to extract estimates of the fractal dimension and lacunarity on 4 x 4 pixel subsets of the image. It was shown that for clouds of fractal dimension 2.7 or less, appropriate thresholding on fractal dimension and lacunarity yielded a 64 x 64 edge-detected image with all or most of the cloud background removed. These images were enhanced by an erosion and dilation to provide the final image passed to the lookup table. While the ultimate goal was to pass the final image to a neural network for identification, this work shows the applicability of fractal geometry to the problems of image segmentation, edge detection and separating a target of interest from a natural background.

Fuller, J. Joseph↗

NASA NTRS2019

SPARROW: A Steam Propelled Autonomous Retrieval Robot for Ocean Worlds

This paper presents the results of a NIAC Phase I study into the use of a propulsively hopping robot for the exploration of Europa’s rugged, icy surface. Named the“Steam Propelled Retrieval Robot for Ocean Worlds,” SPARROW is a multi-thruster robot passively gimballed within a protective, spherical shell, which enables it to freely rotate, self-right, and tumble over chaotic terrains. SPARROW is envisioned as a soccerball-sized payload to a primary lander mission. Europa’s abundant surface ice would be harvested as an in situ propellant source. The principal objective of SPARROW would be to increase the science return of a Europa landed asset by enabling access to distal, spatially distributed geologic units. The design of mobility systems for Europa is challenging, due in part to its almost entirely unconstrained surface topography and strength. Images returned by Voyager and Galileo yielded resolutions on the order of hundreds of meters per pixel, with localized regions reaching 6 meters per pixel—still far larger than a typical rover. A key beneﬁt of SPARROW’s hopping, impact-tolerant design, is that it eliminates the need for a piori information on the terrain topography and surface strength; no surface reaction forces are required for motion. In this context, SPARROW is entirely terrain agnostic. In this paper we detail the results of three study objectives: i) to quantify the energy required to collect surface ice, change its phase, and maintain propellant temperature, ii) to identify control and estimation strategies that enable SPARROW to successfully reach, and return from, regions of scientiﬁc interest, and iii) to characterize the impact of SPARROW’s range on likely science return. Five water-based propellant architectures are presented alongside their mass, power, and volume requirements. Monte Carlo simulations of SPARROW hopping and tumbling over 1 km of glacial ice are summarized, characterizing SPARROW’s sensitivity to uncertainty in: initial conditions, thrust control, and cage-terrain interaction. Finally, a science traceability matrix is presented, which details the effect of sortie range on three science goals: constraining Europa’s evolutionary morphology, assessing sub-surface ocean habitability, and searching for life and/or biosignatures.

Phillips, Cynthia↗

NASA NTRSDate not supplied

Post-Flight Quantification of LOFTID Aeroshell Deflection Using Feature Tracking

The Low-Earth Orbit Flight Test of an Inflatable Decelerator (LOFTID) was a flight demonstration of the Hypersonic Inflatable Aerodynamic Decelerator (HIAD) technology, which has the potential to enable delivery of heavy payloads to Mars, Venus, and Titan, as well as return to Earth. Unlike rigid aeroshells that are constrained by the diameter of the launch vehicle shroud, inflatable aeroshells can be deployed to a much larger drag area, thus allowing a more massive spacecraft to begin its deceleration at higher altitudes and experience less heating. On November 10, 2022, the LOFTID reentry vehicle launched aboard a United Launch Alliance Atlas V rocket to low-Earth orbit. The aeroshell was inflated to its full 6-meter diameter, and the vehicle then successfully re-entered the atmosphere, landing in the Pacific Ocean. The aeroshell was composed of seven tori bound together by high strength straps to create a 70-degree half-angle sphere-cone, and the forebody was covered with a flexible thermal protection system (FTPS) (Fig. 1). The centerbody of the vehicle housed six visual cameras. Each camera was made up of 1920 x 1080 pixels and had a field-of-view (FOV) of 85.4° x 55.6°, resulting in a resolution of less than 0.1" at all locations on the aftbody side of the aeroshell. The approximate locations of the cameras and their associated FOVs is shown in Fig. 2. The high loads experienced during flight resulted in the cone of the aeroshell deflecting. This behavior was seen during the static load testing of the aeroshell in May 2021, in which loads ranging from 1,000 to 20,000 lbf were applied, and deflections of up to ~1.7° were observed. The LOFTID team was interested in estimating the deflection of the aeroshell during the its entry into Earth's atmosphere. Before launch, 1"-diameter black circles were drawn on select structural straps for tracking with the visual cameras; the change in position of these features could then be used to calculate aeroshell deflection angle.

Hannah S. Alpert↗

NASA NTRSDate not supplied

Post-Flight Quantification of LOFTID Aeroshell Deflection Using Feature Tracking

The Low-Earth Orbit Flight Test of an Inflatable Decelerator (LOFTID) was a flight demonstration of the Hypersonic Inflatable Aerodynamic Decelerator (HIAD) technology, which has the potential to enable delivery of heavy payloads to Mars, Venus, and Titan, as well as return to Earth. Unlike rigid aeroshells that are constrained by the diameter of the launch vehicle shroud, inflatable aeroshells can be deployed to a much larger drag area, thus allowing a more massive spacecraft to begin its deceleration at higher altitudes and experience less heating. On November 10, 2022, the LOFTID reentry vehicle launched aboard a United Launch Alliance Atlas V rocket to low-Earth orbit. The aeroshell was inflated to its full 6-meter diameter, and the vehicle then successfully re-entered the atmosphere, landing in the Pacific Ocean. The aeroshell was composed of seven tori bound together by high strength straps to create a 70-degree half-angle sphere-cone, and the forebody was covered with a flexible thermal protection system (FTPS) (Fig. 1). The centerbody of the vehicle housed six visual cameras. Each camera was made up of 1920 x 1080 pixels and had a field-of-view (FOV) of 85.4° x 55.6°, resulting in a resolution of less than 0.1" at all locations on the aftbody side of the aeroshell. The approximate locations of the cameras and their associated FOVs is shown in Fig. 2. The high loads experienced during flight resulted in the cone of the aeroshell deflecting. This behavior was seen during the static load testing of the aeroshell in May 2021, in which loads ranging from 1,000 to 20,000 lbf were applied, and deflections of up to ~1.7° were observed. The LOFTID team was interested in estimating the deflection of the aeroshell during the its entry into Earth's atmosphere. Before launch, 1"-diameter black circles were drawn on select structural straps for tracking with the visual cameras; the change in position of these features could then be used to calculate aeroshell deflection angle.

Hannah S. Alpert↗

DOE OSTI2023

A comparison of past and present computational methods for shape analysis of double-shell x-ray radiographs

Implosion symmetry is a key requirement in achieving a robust burning plasma in inertial confinement fusion experiments. In double-shell capsule implosions, we are interested in the shape of the inner shell as it pushes on the fuel. Shape analysis is a popular technique for studying said symmetry during implosion. Combinations of filtering and contour-finding algorithms are studied for their promise in reliably recovering Legendre shape coefficients from synthetic radiographs of double-shell capsules with applied levels of noise. A radial lineout max(slope) method when used on an image pre-filtered with non-local means and a variant of the marching squares algorithm are able to recover p 0 , p 2 , and p 4 maxslope Legendre shape coefficients with mean pixel discrepancy errors of 2.81 and 3.06, respectively, for the noisy synthetic radiographs we consider. Here, this improves upon prior radial lineout methods paired with Gaussian filtering, which we show to be unreliable and whose performance is dependent on input parameters that are difficult to estimate.

47 OTHER INSTRUMENTATION↗

DOE OSTI2021

Individual Wave Detection and Tracking within a Rotating Detonation Engine through Computer Vision Object Detection applied to High-Speed Images

Known for their simplistic design and continuous detonation, rotating detonation engines (RDEs) constitute a majority of current pressure gain combustion (PGC) research efforts. Experimental RDE operation times have been continuously extended through the use of rig cooling techniques. As the window of observable behavior is expanded, and as the technology matures toward eventual integration within gas turbines, monitoring techniques must evolve to better match industrial diagnostics. High-speed image analysis techniques prove useful to capture and evaluate the unsteady detonation behavior within the RDE. Traditional image analysis techniques, however, require extensive processing times which prohibit simultaneous monitoring. To better address this problem, a computer vision object detection methodology is proposed to quickly detect individual detonation waves within a single down-axis image. Detonation waves are detected in individual images by the implemented computer vision method You Only Look Once (YOLO) object detection network. In order to detect detonation waves, the network must first be trained using RDE images of interest, for which each required phase of network development is outlined. Detection of waves is improved through proper treatment of the collected image set, variation of Intersection over Union (IoU) and confidence thresholding, and through a parametric study of annotation dimensions. Each detected wave is described by its location and rotational direction, and locations are tracked to calculate wave velocity across each frame, leading to a timestep resolution of 20 µs. Wave velocities are also calculated through a series of frames, leading to a suitable average velocity estimation using as few as 10 frames. Uncertainty analysis accounting for variation in camera framerate, pixel width and annotation centroid locations estimates a total uncertainty of ±4.3% for velocity calculations, using the smallest annotation boxes. This method offers great reductions in processing times, as a step toward real-time monitoring of detonation waves within an RDE. Improving on previous studies, this technique is impartial to wave modes not included in the original training set and calculates wave velocities independent of high-speed pressure data. The ability to isolate waves within predicted bounding boxes will likely facilitate analysis of pixel intensity variation as an estimation of wave strength in future work.

Johnson, Kristyn↗

DOE OSTI2021

Individual Wave Detection and Tracking within a Rotating Detonation Engine through Computer Vision Object Detection applied to High-Speed Images

Known for their simplistic design and continuous detonation, rotating detonation engines (RDEs) constitute a majority of current pressure gain combustion (PGC) research efforts. Experimental RDE operation times have been continuously extended through the use of rig cooling techniques. As the window of observable behavior is expanded, and as the technology matures toward eventual integration within gas turbines, monitoring techniques must evolve to better match industrial diagnostics. High-speed image analysis techniques prove useful to capture and evaluate the unsteady detonation behavior within the RDE. Traditional image analysis techniques, however, require extensive processing times which prohibit simultaneous monitoring. To better address this problem, a computer vision object detection methodology is proposed to quickly detect individual detonation waves within a single down-axis image. Detonation waves are detected in individual images by the implemented computer vision method You Only Look Once (YOLO) object detection network. In order to detect detonation waves, the network must first be trained using RDE images of interest, for which each required phase of network development is outlined. Detection of waves is improved through proper treatment of the collected image set, variation of Intersection over Union (IoU) and confidence thresholding, and through a parametric study of annotation dimensions. Each detected wave is described by its location and rotational direction, and locations are tracked to calculate wave velocity across each frame, leading to a timestep resolution of 20 µs. Wave velocities are also calculated through a series of frames, leading to a suitable average velocity estimation using as few as 10 frames. Uncertainty analysis accounting for variation in camera framerate, pixel width and annotation centroid locations estimates a total uncertainty of ±4.3% for velocity calculations, using the smallest annotation boxes. This method offers great reductions in processing times, as a step toward real-time monitoring of detonation waves within an RDE. Improving on previous studies, this technique is impartial to wave modes not included in the original training set and calculates wave velocities independent of high-speed pressure data. The ability to isolate waves within predicted bounding boxes will likely facilitate analysis of pixel intensity variation as an estimation of wave strength in future work.

Johnson, Kristyn↗