Search NASA⌕ Search

SEARCH · Search NASA

Results for “Vision”

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 289 records · Page 16
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Sequence length scaling in vision transformers for scientific images on frontier

Vision Transformers (ViTs) are pivotal for foundational models in scientific imagery, including Earth science applications, due to their capability to process large sequence lengths. While transformers for text have inspired scaling sequence lengths in ViTs, adapting these for ViTs introduces unique challenges. We develop distributed sequence parallelism for ViTs, enabling them to handle up to 1M tokens. Our approach, leveraging DeepSpeed-Ulysses and Long-Sequence-Segmentation with model sharding, is the first to apply sequence parallelism in ViT training, achieving a 94% batch scaling efficiency on 2,048 AMD-MI250X GPUs. Evaluating sequence parallelism in ViTs, particularly in models up to 10B parameters, highlighted substantial bottlenecks. We countered these with hybrid sequence, pipeline, and flash attention strategies, to scale beyond single GPU memory limits. Our method significantly enhances climate modeling accuracy by 20% in temperature predictions, marking the first training of a vision transformer model to convergence with a sequence length of 188K tokens, using full self-attention.

Tsaris, Aristeidis (aris) [ORNL] (ORCID:0000000277↗
DOE OSTI2020

Photon Irradiation Effects on the Enduray Vision System for Nuclear Inspections

Video cameras with greater radiation hardening are needed for the operation and remote handling of equipment in nuclear reactor inspection and refueling applications. Vega Wave Systems has developed a radiation-hardened vision system for nuclear energy applications. The program described in this report provided radiation-hardness testing of this vision system using the Argonne Low-Energy Accelerator Facility (LEAF). This report presents the results of irradiation tests performed on several of Vega Wave System’s cameras at dose rates ranging from 1 kGy/hr up to 10 kGy/hr and up to total doses of >500 kGy. The results were excellent, providing proof of the camera system’s immunity to radiation effects at all levels tested.

36 MATERIALS SCIENCE↗
DOE OSTI2017

Storage 2020: A Vision for the Future of HPC Storage

The explosive growth in data over the next five years that will accompany exascale simulations and new experimental detectors will enable new data-driven science across virtually every domain. At the same time, new nonvolatile storage technologies will enter the market in volume and upend long-held principles used to design the storage hierarchy. The disruption that these forces will bring to bear on high-performance computing (HPC) will also create significant opportunities to innovate and accelerate scientific discovery. To ensure that NERSC fully capitalizes on these opportunities, we have developed a comprehensive vision for the future of storage in HPC and identified short- and long-term strategic goals to effectively realize this vision. This report presents the results of this effort and offers a blueprint for designing a storage infrastructure for supporting HPC through 2025 and beyond.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Development of Transmission Measurements Capability at VISION Spectrometer [Slides]

This presentation covers the development of transmission measurements capability at VISION Spectrometer. Main takeaways include that INS and transmission measurements are and should be the most significant part of TSL validation, as they are the fundamental quantities used by Monte Carlo neutronics codes. Additionally, the development of transmission capability at VISION instrument has commenced. Promising first results have been obtained with a big impact by the use of the collimators. Finally, Areas of improvement have been identified and new measurements are planned for June.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2021

Considerations regarding the Use of Computer Vision Machine Learning in Safety-Related or Risk-Significant Applications in Nuclear Power Plants

With the advancements made to date in the field of artificial intelligence (AI), significant potential exists to utilize AI capabilities for nuclear power plant (NPP) applications. AI can replicate human decision making and it is usually faster and more accurate than humans. For implementations that impact critical NPP applications (e.g., safety-related or non-safety systems that potentially affect overall plant risk), a deeper safety analysis of the AI methods is necessary. AI applied to NPP operations could resemble the use of digital I&C (DI&C) because such applications involve digital computer hardware and custom-designed software that input plant data, execute complex software algorithms, and output the results to a system or licensed human operator to potentially provoke an action. For AI methods to be compliant with current safety requirements for DI&C, AI compatibility must be evaluated, and AI-related gaps may exist that prevent the prompt deployment of AI in NPPs. This effort aims to evaluate how example AI technologies align with the DI&C safety framework, and discusses how they could be analyzed, modeled, tested, and validated in a manner similar to typical DI&C technologies. Because AI is a broad field that encompasses areas such as machine learning (ML), natural language processing, and computer vision, this research focused on a subset of methods categorized as the computer vision ML (CVML) methods. This report explores two CVML use cases, gauge reading and fire watch, considered relevant to the DI&C standards, as they could play a safety-critical role. For the gauge reading use case, a CVML-enabled technology that can read gauges at oblique angles is utilized. For the fire watch use case, a CVML-enabled technology is utilized that migrates fire watch from a manual (human) approach to automated fire detection. These use cases are mainly intended to give context to the CVML system discussion. This effort assumes the worst-case scenario, with the CVML system being used to replace a safety-related or risk-significant system, thus requiring evaluation. Evaluating CVML against most of the relevant safety requirements for DI&C yielded several CVML-specific considerations due to the uniqueness of its characteristics in comparison with typical DI&C systems. For example, CVML models often employ commonly used (open-source) datasets, and it is not always possible to determine the level of overlap among open-source datasets. Therefore, the independence of the developed CVML models when demonstrating diversity is questionable, therefore creating vulnerability to common cause failure (CCF). The design verification process is also impacted since the data overlap could result in overestimation of the software validation and verification (V&V) performance results. Section 2 of this report evaluates a list of the identified CVML-specific characteristics and discusses the resulting considerations and potential solutions in the context of each referenced requirement. A summation is provided in Section 3. This report is not to be used as a guideline. It was developed to identify and consider issues in the implementation of ML technologies used to augment activities that may have a bearing on plant operation. The report draws parallels to the use of DI&C technologies, for which many standards are available to guide their use in nuclear plant operation. It considers the technologies and some of the potential implications of their use in safety-related applications but is not intended to address regulatory or licensing related issues.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2022

DOE OE 2021 Strategy White Papers on Microgrids: Program Vision, Objectives, and R&D Targets in 5 and 10 years–Topic Area #1

This white paper describes the program vision, objectives, and R&D targets in 5 to 10 years for the Department of Energy (DOE) Office of Electricity (OE) Microgrid R&D Program. The vision is to facilitate the nation’s transitions to (1) a more resilient and reliable, (2) more decarbonized electricity infrastructure, in which (3) microgrids have a reduced cost to implement. This strategy is developed in the context that the United States’ electricity delivery system is becoming more distributed in nature. The electricity generation capacity in 10 years may be 30-50% distributed energy assets.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

Visioning Energy: Science Fiction Author-Energy Researcher Collaboration Workshop Recap

The Visioning Energy: Science Fiction Author-Energy Researcher Collaboration Workshop brought together speculative fiction authors and NREL researchers to examine possible scenarios of the future. The goals of the workshop were to support out-of-the-box thinking and creative future visioning for the researchers and to provide authors with insights into the latest clean energy technologies and their potential. This report outlines the proceedings of the workshop and insights from the collaborative brainstorming activity, highlighting themes and opportunities for further exploration.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2025

Integrating Machine-learning-assisted Computer Vision with RICH System

Developments in artificial intelligence have vastly expanded the capabilities of robots. Currently, the Spallation Neutron Source (SNS) beamlines at Oak Ridge National Lab (ORNL) have robotic sample loaders to increase the efficiency of running experiments. However, they require retraining if anything about the situation changes, e.g., where the samples are, and cannot notice if errors occur. So, the viability of using computer vision and machine learning to enhance these sample loaders’ functionality was investigated. In this project, the RICH system with a Dobot CR3 6-axis robot present at the VULCAN beamline assisted by an Intel Realsense D435i camera, a unique camera that enables convenient translation of 2D pixel coordinates to 3D world points, was programmed to load ceramic crucibles into a thermogravimetric analyzer (TGA) furnace. An algorithm was constructed in Python with three major phases planned: (1) obtaining a sample, (2) moving it to the target location, and then (3) bringing the sample back to its original location once the experiment finished. In the first phase, the algorithm would dynamically detect sample locations using ArUco markers to recognize the samples’ general location and a custom-trained yolov5 object detection model to locate the crucibles’ centers. Afterward, the robot would be directed to pick up samples based on the crucibles’ calculated positions. In the second phase, the robot would move the sample to a secondary point, reorient its grip, and place the sample at the target location. In the final phase, the robot would determine whether the sample was intact and would bring it back to its original place if it was or raise an alarm. Using this algorithm, the robot was able to pick up different types of crucibles at varying positions. These results indicate that integrating machine-learning-assisted computer vision with robotic sample loaders can result in effective autonomous detection of samples.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Vision 0.3.2 Users Guide: Revision 0

Vision is a research software application developed by Pacific Northwest National Laboratory (PNNL) that incorporates the scientific and operational expertise for reviewing data from treaty monitoring radionuclide stations. These stations are part of a worldwide network to monitor for nuclear explosions, and the data they produce are critical to make the determination of whether a sample is from a nuclear explosion or some other source (i.e., nuclear reactor or medical isotope production facility). Stations deliver their measurements and system status to the International Monitoring System (IMS), which forwards it via email to all subscribers. Vision provides tools for viewing and analyzing these stations’ state of health data.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Permeability Prediction Using Vision Transformers

Accurate permeability predictions remain pivotal for understanding fluid flow in porous media, influencing crucial operations across petroleum engineering, hydrogeology, and related fields. Traditional approaches, while robust, often grapple with the inherent heterogeneity of reservoir rocks. With the advent of deep learning, convolutional neural networks (CNNs) have emerged as potent tools in image-based permeability estimation, capitalizing on micro-CT scans and digital rock imagery. This paper introduces a novel paradigm, employing vision transformers (ViTs)—a recent advancement in computer vision—for this crucial task. ViTs, which segment images into fixed-sized patches and process them through transformer architectures, present a promising alternative to CNNs. We present a methodology for implementing ViTs for permeability prediction, its results on diverse rock samples, and a comparison against conventional CNNs. The prediction results suggest that, with adequate training data, ViTs can match or surpass the predictive accuracy of CNNs, especially in rocks exhibiting significant heterogeneity. This study underscores the potential of ViTs as an innovative tool in permeability prediction, paving the way for further research and integration into mainstream reservoir characterization workflows.

58 GEOSCIENCES↗
DOE OSTI2025

ESPPU INPUT: C$^3$ within the "Linear Collider Vision"

The Linear Collider Vision calls for a Linear Collider Facility with a physics reach from a Higgs Factory to the TeV-scale with $e^+e^{-}$ collisions. One of the technologies under consideration for the accelerator is a cold-copper distributed-coupling linac capable of achieving high gradient. This technology is being pursued by the C$^3$ collaboration to understand its applicability to future colliders and broader scientific applications. In this input we share the baseline parameters for a C$^3$ Higgs-factory and the energy reach of up to 3 TeV in the 33 km tunnel foreseen under the Linear Collider Vision. Recent results, near-term plans and future R&D needs are highlighted.

43 PARTICLE ACCELERATORS↗
NASA NTRS1973

Effect of elimination of nitrogen and/or hypoxia or restricted visual environment on color vision and range of accommodation

The effects upon range of accommodation and color vision of reduced atmospheric pressure, at partial and complete elimination of nitrogen, of hypoxia, and of exposure for varying periods of time to restricted visual environment, have been studied alone or in various combinations. Measurements were made on the electroretinogram, the electrooculogram, and the diameter of the retinal vessels as an indicator of blood flow to the retina at the time of total elimination of nitrogen. An objective method was used to test range of accommodation. In the color vision test the flicker colors of a Benham's top were matched with a colorimeter.

Wolbarsht, M. L.↗
NASA NTRS1975

Aircraft cockpit vision: Math model

A mathematical model was developed to describe the field of vision of a pilot seated in an aircraft. Given the position and orientation of the aircraft, along with the geometrical configuration of its windows, and the location of an object, the model determines whether the object would be within the pilot's external vision envelope provided by the aircraft's windows. The computer program using this model was implemented and is described.

Bashir, J.↗
NASA NTRS1975

Perception of linear horizontal self-motion induced by peripheral vision /linearvection/ - Basic characteristics and visual-vestibular interactions

The basic characteristics of the sensation of linear horizontal motion have been studied. Objective linear motion was induced by means of a moving cart. Visually induced linear motion perception (linearvection) was obtained by projection of moving images at the periphery of the visual field. Image velocity and luminance thresholds for the appearance of linearvection have been measured and are in the range of those for image motion detection (without sensation of self motion) by the visual system. Latencies of onset are around 1 sec and short term adaptation has been shown. The dynamic range of the visual analyzer as judged by frequency analysis is lower than the vestibular analyzer. Conflicting situations in which visual cues contradict vestibular and other proprioceptive cues show, in the case of linearvection a dominance of vision which supports the idea of an essential although not independent role of vision in self motion perception.

Berthoz, A.↗
NASA NTRS1979

Contribution to the theory of photopic vision: Retinal phenomena

Principles of thermodynamics are applied to the study of the ultramicroscopic anatomy of the inner eye. Concepts introduced and discussed include: the retina as a three-dimensional sensor, light signals as coherent beams in relation to the dimensions of retinal pigments, pigment effects topographed by the conjugated antennas effect, visualizing lights, the autotropic function of hemoglobin and some cytochromes, and reversible structural arrangements during photopic adaptation. A paleoecological diagram is presented which traces the evolution of scotopic vision (primitive system) to photopic vision (secondary system) through the emergence of structures sensitive to the intensity, temperature, and wavelengths of the visible range.

Calvet, H.↗
NASA NTRS1981

Marking parts to aid robot vision

The premarking of parts for subsequent identification by a robot vision system appears to be beneficial as an aid in the automation of certain tasks such as construction in space. A simple, color coded marking system is presented which allows a computer vision system to locate an object, calculate its orientation, and determine its identity. Such a system has the potential to operate accurately, and because the computer shape analysis problem has been simplified, it has the ability to operate in real time.

Bales, J. W.↗
NASA NTRS1982

An overview of computer vision

An overview of computer vision is provided. Image understanding and scene analysis are emphasized, and pertinent aspects of pattern recognition are treated. The basic approach to computer vision systems, the techniques utilized, applications, the current existing systems and state-of-the-art issues and research requirements, who is doing it and who is funding it, and future trends and expectations are reviewed.

Gevarter, W. B.↗
NASA NTRS1984

Edge analyzing properties of center/surround response functions in cybernetic vision

The ability of center/surround response functions to make explicit high resolution spatial information in optical images was investigated by performing convolutions of two dimensional response functions and image intensity functions (mainly edges). The center/surround function was found to have the unique property of separating edge contrast from shape variations and of providing a direct basis for determining contrast and subsequently shape of edges in images. Computationally simple measures of contrast and shape were constructed for potential use in cybernetic vision systems. For one class of response functions these measures were found to be reasonably resilient for a range of scan direction and displacements of the response functions relative to shaped edges. A pathological range of scan directions was also defined and methods for detecting and handling these cases were developed. The relationship of these results to biological vision is discussed speculatively.

Jobson, D. J.↗