Search NASA⌕ Search

SEARCH · Search NASA

Results for “Overhead Image”

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 109 records · Page 6

Comparing Deep Learning Approaches for Understanding Genotype × Phenotype Interactions in Biomass Sorghum

We explore the use of deep convolutional neural networks (CNNs) trained on overhead imagery of biomass sorghum to ascertain the relationship between single nucleotide polymorphisms (SNPs), or groups of related SNPs, and the phenotypes they control. We consider both CNNs trained explicitly on the classification task of predicting whether an image shows a plant with a reference or alternate version of various SNPs as well as CNNs trained to create data-driven features based on learning features so that images from the same plot are more similar than images from different plots, and then using the features this network learns for genetic marker classification. We characterize how efficient both approaches are at predicting the presence or absence of a genetic markers, and visualize what parts of the images are most important for those predictions. We find that the data-driven approaches give somewhat higher prediction performance, but have visualizations that are harder to interpret; and we give suggestions of potential future machine learning research and discuss the possibilities of using this approach to uncover unknown genotype × phenotype relationships.

Zhang, Zeyu↗

System Would Generate Virtual Heads-Up Display

Proposed helmet-mounted electronic display system superimposes full-color alphanumerical and/or graphical information onto observer's visual field. Displayed information projected directly onto observer's retinas, giving observer illusion of full-size computer display in foreground or background. Display stereoscopic, holographic, or in form of virtual image. Used by pilots to view navigational information while looking outside or at instruments, by security officers to view information about critical facilities while looking at visitors, or possibly even stock-exchange facilities to view desktop monitors and overhead displays simultaneously. System includes acousto-optical tunable filter (AOTF), which acts as both spectral filter and spatial light modulator.

Lambert, James L.↗

Analytic variance estimates of Swank and Fano factors

Purpose: Variance estimates for detector energy resolution metrics can be used as stopping criteria in Monte Carlo simulations for the purpose of ensuring a small uncertainty of those metrics and for the design of variance reduction techniques. Methods: The authors derive an estimate for the variance of two energy resolution metrics, the Swank factor and the Fano factor, in terms of statistical moments that can be accumulated without significant computational overhead. The authors examine the accuracy of these two estimators and demonstrate how the estimates of the coefficient of variation of the Swank and Fano factors behave with data from a Monte Carlo simulation of an indirect x-ray imaging detector. Results: The authors' analyses suggest that the accuracy of their variance estimators is appropriate for estimating the actual variances of the Swank and Fano factors for a variety of distributions of detector outputs. Conclusions: The variance estimators derived in this work provide a computationally convenient way to estimate the error or coefficient of variation of the Swank and Fano factors during Monte Carlo simulations of radiation imaging systems.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Parallel implementation and evaluation of motion estimation system algorithms on a distributed memory multiprocessor using knowledge based mappings

Several techniques to perform static and dynamic load balancing techniques for vision systems are presented. These techniques are novel in the sense that they capture the computational requirements of a task by examining the data when it is produced. Furthermore, they can be applied to many vision systems because many algorithms in different systems are either the same, or have similar computational characteristics. These techniques are evaluated by applying them on a parallel implementation of the algorithms in a motion estimation system on a hypercube multiprocessor system. The motion estimation system consists of the following steps: (1) extraction of features; (2) stereo match of images in one time instant; (3) time match of images from different time instants; (4) stereo match to compute final unambiguous points; and (5) computation of motion parameters. It is shown that the performance gains when these data decomposition and load balancing techniques are used are significant and the overhead of using these techniques is minimal.

Choudhary, Alok Nidhi↗

Scalable Transparent Checkpoint-Restart of Global Address Space Applications on Virtual Machines over Infiniband

Checkpoint-Restart is one of the most used software approaches to achieve fault-tolerance in high-end clusters. While standard techniques typically focus on user-level solutions, the advent of virtualization software has enabled efficient and transparent system-level approaches. In this paper, we present a scalable transparent system-level solution to address fault-tolerance for applications based on global address space (GAS) programming models on Infiniband clusters. In addition to handling communication, the solution addresses transparent checkpoint of user-generated files. We exploit the support for the Infiniband network in the Xen virtual machine environment. We have developed a version of the Aggregate Remote Memory Copy Interface (ARMCI) one-sided communication library capable of suspending and resuming applications. We present efficient and scalable mechanisms to distribute checkpoint requests and to backup virtual machines memory images and file systems. We tested our approach in the context of NWChem, a popular computational chemistry suite. We demonstrated that NWChem can be executed, without any modification to the source code, on a virtualized 8-node cluster with very little overhead (below 3%). We observe that the total checkpoint time is limited by disk I/O. Finally, we measured system-size depended components of the checkpoint time on up to 1024 cores (128 nodes), demonstrating the scalability of our approach in medium/large-scale systems.

97 MATHEMATICS AND COMPUTING↗

Out-of-distribution detection with non-parametric density estimation for models predicting processing history of uranium ore concentrates

The rapid advancement in machine learning (ML) and computer vision (CV) coincides with the growth of interest in deploying these ML/CV models in numerous fields from medicine to social science. Similar to those areas, we have witnessed a great number of works in materials science employing ML/CV models – neural networks in particular – in their studies in recent years. These models have proven to obtain accurate performance in various tasks. However, these models struggle to attain a similar performance when encountering test samples coming from a distribution that is different from the training set. More importantly, they fail without providing any warning to the users. Therefore, we propose a framework for detecting out-of-distribution (OOD) samples to alert users when a human intervention might be necessary in this work. Specifically, we explore the use of a non-parametric density estimation method to detect OOD samples. Here, we assess OOD detection capability of the proposed framework on ML models developed for categorizing precipitation routes of U 3 O 8 when encountering OOD datasets that contain samples (1) undergone different imaging acquisition process, (2) undergone different material synthesis process, and (3) different materials than ID set. Through those experiments, we achieve an average area under the receiver operating characteristic (AUROC) of at least 91% on average in detecting OOD samples. With minimal overhead cost and superior performance, the proposed framework enables a reliable and safe system when deploying in real-world scenarios.

Convolutional neural networks↗

Physical Validation of TRMM TMI and PR Monthly Rain Products Over Oklahoma

The Tropical Rainfall Measuring Mission (TRMM) provides monthly rainfall estimates using data collected by the TRMM satellite. These estimates cover a substantial fraction of the earth's surface. The physical validation of TRMM estimates involves corroborating the accuracy of spaceborne estimates of areal rainfall by inferring errors and biases from ground-based rain estimates. The TRMM error budget consists of two major sources of error: retrieval and sampling. Sampling errors are intrinsic to the process of estimating monthly rainfall and occur because the satellite extrapolates monthly rainfall from a small subset of measurements collected only during satellite overpasses. Retrieval errors, on the other hand, are related to the process of collecting measurements while the satellite is overhead. One of the big challenges confronting the TRMM validation effort is how to best estimate these two main components of the TRMM error budget, which are not easily decoupled. This four-year study computed bulk sampling and retrieval errors for the TRMM microwave imager (TMI) and the precipitation radar (PR) by applying a technique that sub-samples gauge data at TRMM overpass times. Gridded monthly rain estimates are then computed from the monthly bulk statistics of the collected samples, providing a sensor-dependent gauge rain estimate that is assumed to include a TRMM equivalent sampling error. The sub-sampled gauge rain estimates are then used in conjunction with the monthly satellite and gauge (without sub- sampling) estimates to decouple retrieval and sampling errors. The computed mean sampling errors for the TMI and PR were 5.9% and 7.796, respectively, in good agreement with theoretical predictions. The PR year-to-year retrieval biases exceeded corresponding TMI biases, but it was found that these differences were partially due to negative TMI biases during cold months and positive TMI biases during warm months.

Fisher, Brad L.↗

Tracking particles by passing messages between images

Methods to extract information from the tracking of mobile objects/particles have broad interest in biological and physical sciences. Techniques based on the simple criterion of proximity in time-consecutive snapshots are useful to identify the trajectories of the particles. However, they become problematic as the motility and/or the density of the particles increases because of the uncertainties on the trajectories that particles have followed during the acquisition time of the images. Here, we report efficient methods for learning parameters of the dynamics of the particles from their positions in time-consecutive images. Our algorithm belongs to the class of message-passing algorithms, also known in computer science, information theory and statistical physics under the name of Belief Propagation (BP). The algorithm is distributed, thus allowing parallel implementation suitable for computations on multiple machines without significant inter-machine overhead. We test our method on the model example of particle tracking in turbulent flows, which is particularly challenging due to the strong transport that those flows produce. Our numerical experiments show that the BP algorithm compares in quality with exact Markov Chain Monte-Carlo algorithms, yet BP is far superior in speed. We also suggest and analyze a random-distance model that provides theoretical justification for BP accuracy. Methods developed here systematically formulate the problem of particle tracking and provide fast and reliable tools for its extensive range of applications.

99 GENERAL AND MISCELLANEOUS↗

From computer images to video presentation: Enhancing technology transfer

With NASA placing increased emphasis on transferring technology to outside industry, NASA researchers need to evaluate many aspects of their efforts in this regard. Often it may seem like too much self-promotion to many researchers. However, industry's use of video presentations in sales, advertising, public relations and training should be considered. Today, the most typical presentation at NASA is through the use of vu-graphs (overhead transparencies) which can be effective for text or static presentations. For full blown color and sound presentations, however, the best method is videotape. In fact, it is frequently more convenient due to its portability and the availability of viewing equipment. This talk describes techniques for creating a video presentation through the use of a combined researcher and video professional team.

Beam, Sherilee F.↗

A Summary of Results from Two Full-Scale Fokker F28 Fuselage Section Drop Tests

During 2017, two vertical drop tests were conducted on two partial sections removed from a Fokker F28 MK4000 aircraft as a part of a joint NASA / Federal Aviation Administration (FAA) effort to investigate the crashworthiness characteristics of Transport Category Aircraft, as defined by 14 Code of Federal Regulations, Part 25. The first test was a pure vertical drop test of a relatively uniform forward section, which included an underfloor area for luggage. The second test was a canted drop test onto a sloped surface of a portion of the fuselage representing the wingbox stiffened structure. In both tests, accelerometers were installed on the floor, seat track, luggage, and overhead bin to measure responses in the two airframe sections. In addition, ten Anthropomorphic Test Devices (ATDs, a.k.a. crash test dummies) were included in each test to measure the potential of onboard occupant injury. Self-contained data recorders, logging accelerations and rotational rates, were also used on the seat tracks and lower structure for evaluation as potential crash recording devices in future tests. Finally, the starboard side of each section was painted with a stochastic black and white speckle pattern for use in full field photogrammetric imaging techniques. The results collected show notable differences in the Forward and Wingbox Section responses. The Forward Section floor accelerations showed a relatively uniform response of approximately 7 g throughout the impact event. This section exhibited large amounts of subfloor crushing, floor stiffener failures and seat deformation upon impact. These results are contrasted by the Wingbox Section accelerations, which showed large differences when comparing accelerations recorded from either the rear or the front portion of the section reaching peaks of 39 g in some locations. Additionally, the Wingbox Section test induced a forward motion caused from the rotation at impact. With the exception of the lower cavity, there was minimal deformation in the Wingbox Section, and ATD responses were consistently higher than those for the Forward Section. A complete set of results are presented for the airframe, seat, ATD, overhead bin, and subfloor regions for each section. The ATD results are compared to current injury criteria, and determinations will be made on the likelihood of injury. Additional results for cargo-hold stored luggage are presented in an attempt to provide a component-level characterization for better understanding of under-floor loading. Finally, a discussion of the relevance of the results for a proposed airframe level crashworthiness guideline are presented.

Littell, Justin D.↗

Computing quality scores and uncertainty for approximate pattern matching in geospatial semantic graphs

Geospatial semantic graphs provide a robust foundation for representing and analyzing remote sensor data. In particular, they support a variety of pattern search operations that capture the spatial and temporal relationships among the objects and events in the data. However, in the presence of large data corpora, even a carefully constructed search query may return a large number of unintended matches. This work considers the problem of calculating a quality score for each match to the query, given that the underlying data are uncertain. As a result, we present a preliminary evaluation of three methods for determining both match quality scores and associated uncertainty bounds, illustrated in the context of an example based on overhead imagery data.

97 MATHEMATICS AND COMPUTING↗

Efficient distributed continual learning for steering experiments in real-time

Deep learning has emerged as a powerful method for extracting valuable information from large volumes of data. However, when new training data arrives continuously (i.e., is not fully available from the beginning), incremental training suffers from catastrophic forgetting (i.e., new patterns are reinforced at the expense of previously acquired knowledge). Training from scratch each time new training data becomes available would result in extremely long training times and massive data accumulation. Rehearsal-based continual learning has shown promise for addressing the catastrophic forgetting challenge, but research to date has not addressed performance and scalability. To fill this gap, we propose an approach based on a distributed rehearsal buffer that efficiently complements data-parallel training on multiple GPUs to achieve high accuracy, short runtime, and scalability. It leverages a set of buffers (local to each GPU) and uses several asynchronous techniques for updating these local buffers in an embarrassingly parallel fashion, all while handling the communication overheads necessary to augment input minibatches using unbiased, global sampling. We further propose a generalization of rehearsal buffers to support both classification and generative learning tasks, as well as more advanced rehearsal strategies (notably Dark Experience Replay, leveraging knowledge distillation). We illustrate this approach with a real-life HPC streaming application from the domain of ptychographic image reconstruction. Furthermore, we run extensive experiments on up to 128 GPUs of the ThetaGPU supercomputer to compare our approach with baselines representative of training-from-scratch (the upper bound in terms of accuracy) and incremental training (the lower bound). Results show that rehearsal-based continual learning achieves a top-5 validation accuracy close to the upper bound, while simultaneously exhibiting a runtime close to the lower bound.

Asynchronous data management↗

Full-Scale Drop Test of a Fokker F28 Wingbox Fuselage Section

During the summer of 2017, a vertical drop test was conducted on a partial section of a Fokker F28 MK4000 aircraft as a part of a joint NASA/FAA effort to investigate the performance of transport category aircraft under realistic crash conditions. Ten Anthropomorphic Test Devices (ATDs, a.k.a. crash test dummies) ranging from 5th to 95th percentile sizes were used for the collection and comparison of occupant loads. Additionally, overhead bin mass simulators were added to achieve a realistic fuselage configuration. The section was dropped with a downward facing pitch angle onto a sloping soil surface in order to simulate a local horizontal velocity in the airframe. Instrumentation consisting of accelerometers was installed to measure floor, seat track, ATD, and overhead bin acceleration responses. Self-contained data recorders logging accelerations and rotational rates were also used on the seat tracks and lower structure as evaluations for crash recording devices in potential future use cases. The right side of the section was painted with a stochastic black and white speckle pattern for use in full field photogrammetric imaging techniques. Results collected from the airframe accelerometers will be presented, and deformation and failures of the test article structure will be discussed. Finally, an examination of the test article motion will be presented using derived components of local velocities with their effect on the impact acceleration and airframe response.

Littell, J. D.↗

Hybrid Checkpointing for MPI Jobs in HPC Environments

As the core count in high-performance computing systems keeps increasing, faults are becoming common place. Check pointing addresses such faults but captures full process images even though only a subset of the process image changes between checkpoints. We have designed a hybrid check pointing technique for MPI tasks of high-performance applications. This technique alternates between full and incremental checkpoints: At incremental checkpoints, only data changed since the last checkpoint is captured. Our implementation integrates new BLCR and LAM/MPI features that complement traditional full checkpoints. This results in significantly reduced checkpoint sizes and overheads with only moderate increases in restart overhead. After accounting for cost and savings, benefits due to incremental checkpoints are an order of magnitude larger than overheads on restarts. We further derive qualitative results indicating an optimal balance between full/incremental checkpoints of our novel approach at a ratio of 1:9, which outperforms both always-full and always-incremental check pointing.

Wang, Chao↗

SCIENCE PARAMETRICS FOR MISSIONS TO SEARCH FOR EARTH-LIKE EXOPLANETS BY DIRECT IMAGING

We use N{sub t} , the number of exoplanets observed in time t, as a science metric to study direct-search missions like Terrestrial Planet Finder. In our model, N has 27 parameters, divided into three categories: 2 astronomical, 7 instrumental, and 18 science-operational. For various ''27-vectors'' of those parameters chosen to explore parameter space, we compute design reference missions to estimate N{sub t} . Our treatment includes the recovery of completeness c after a search observation, for revisits, solar and antisolar avoidance, observational overhead, and follow-on spectroscopy. Our baseline 27-vector has aperture D = 16 m, inner working angle IWA = 0.039'', mission time t = 0-5 yr, occurrence probability for Earth-like exoplanets η = 0.2, and typical values for the remaining 23 parameters. For the baseline case, a typical five-year design reference mission has an input catalog of ∼4700 stars with nonzero completeness, ∼1300 unique stars observed in ∼2600 observations, of which ∼1300 are revisits, and it produces N {sub 1} ∼ 50 exoplanets after one year and N {sub 5} ∼ 130 after five years. We explore offsets from the baseline for 10 parameters. We find that N depends strongly on IWA and only weakly on D. It also depends only weakly on zodiacal light for Z < 50 zodis, end-to-end efficiency for h > 0.2, and scattered starlight for ζ < 10{sup –10}. We find that observational overheads, completeness recovery and revisits, solar and antisolar avoidance, and follow-on spectroscopy are all important factors in estimating N.

79 ASTRONOMY AND ASTROPHYSICS↗

POLAR-UVI and other Coordinated Observations of a Traveling Convection Vortex Event Observed on 24 July 1996

Coordinated analysis of data from the POLAR UVI instrument, ground magnetometers, incoherent scatter radar, solar wind monitors IMP-8 and WIND, and DMSP satellite is focused on a traveling convection vortex (TCV) event on 24 July 1966. Starting at 10:48 UT, ground magnetometers in Greenland and eastern Canada measure pulsations consistent with the passing overhead of a series of alternating TCV field-aligned current pairs. Sondrestrom incoherent scatter radar measures strong modulation of the strength and direction of ionospheric plasma flow, The magnetometer pulsations grow in magnitude over the next hour, peaking in intensity at 11:39 UT, at which time the UVI instrument measures a localized intensification of auroral emissions over central and western Greenland. Subsequent images show the intensification grow in strength and propagate westward (tailward) until approximately 11:58 UT at which time the emissions fade. These observations are consistent with the westward passage of two pairs of moderately intense TCVs over central Greenland followed by a third very intense TCV pair. The intensification of auroral emissions at 11:39 UT is associated with the trailing vortex of the third TCV pair, thought to be the result of an upward field-aligned current. Measurements of the solar wind suggest that a pressure change may be responsible for triggering the first two pairs of TCVS, and that a subsequent sudden change in orientation of the IMF may have produced the intensification of the third TCV pair and the associated aurora] brightening. DMSP particle data indicate that the TCVs occur on field lines which map to the boundary plasma sheet or outer edge of the low latitude boundary layer.

Clauer, C. R.↗

Resolving local ambiguity using semantics of shape.

We demonstrate a new semantic method for automatic analysis of wide-area, high-resolution overhead imagery to tip and cue human intelligence analysts to human activity. In the open demonstration, we find and trace cars and rooftops. Our methodology, extended to analysis of voxels, may be applicable to understanding morphology and to automatic tracing of neurons in large-scale, serial-section TEM datasets. We defined an algorithm and software implementation that efficiently finds all combinations of image blobs that satisfy given shape semantics, where image blobs are formed as a general-purpose, first step that 'oversegments' image pixels into blobs of similar pixels. We will demonstrate the remarkable power (ROC) of this combinatorial-based work flow for automatically tracing any automobiles in a scene by applying semantics that require a subset of image blobs to fill out a rectangular shape, with width and height in given intervals. In most applications we find that the new combinatorial-based work flow produces alternative (overlapping) tracings of possible objects (e.g. cars) in a scene. To force an estimation (tracing) of a consistent collection of objects (cars), a quick-and-simple greedy algorithm is often sufficient. We will demonstrate a more powerful resolution method: we produce a weighted graph from the conflicts in all of our enumerated hypotheses, and then solve a maximal independent vertex set problem on this graph to resolve conflicting hypotheses. This graph computation is almost certain to be necessary to adequately resolve multiple, conflicting neuron topologies into a set that is most consistent with a TEM dataset.

99 GENERAL AND MISCELLANEOUS↗

WE-AB-303-09: Rapid Projection Computations for On-Board Digital Tomosynthesis in Radiation Therapy

Purpose: To facilitate fast and accurate iterative volumetric image reconstruction from limited-angle on-board projections. Methods: Intrafraction motion hinders the clinical applicability of modern radiotherapy techniques, such as lung stereotactic body radiation therapy (SBRT). The LIVE system may impact clinical practice by recovering volumetric information via Digital Tomosynthesis (DTS), thus entailing low time and radiation dose for image acquisition during treatment. The DTS is estimated as a deformation of prior CT via iterative registration with on-board images; this shifts the challenge to the computational domain, owing largely to repeated projection computations across iterations. We address this issue by composing efficient digital projection operators from their constituent parts. This allows us to separate the static (projection geometry) and dynamic (volume/image data) parts of projection operations by means of pre-computations, enabling fast on-board processing, while also relaxing constraints on underlying numerical models (e.g. regridding interpolation kernels). Further decoupling the projectors into simpler ones ensures the incurred memory overhead remains low, within the capacity of a single GPU. These operators depend only on the treatment plan and may be reused across iterations and patients. The dynamic processing load is kept to a minimum and maps well to the GPU computational model. Results: We have integrated efficient, pre-computable modules for volumetric ray-casting and FDK-based back-projection with the LIVE processing pipeline. Our results show a 60x acceleration of the DTS computations, compared to the previous version, using a single GPU; presently, reconstruction is attained within a couple of minutes. The present implementation allows for significant flexibility in terms of the numerical and operational projection model; we are investigating the benefit of further optimizations and accurate digital projection sub-kernels. Conclusion: Composable projection operators constitute a versatile research tool which can greatly accelerate iterative registration algorithms and may be conducive to the clinical applicability of LIVE. National Institutes of Health Grant No. R01-CA184173; GPU donation by NVIDIA Corporation.

60 APPLIED LIFE SCIENCES↗