Search NASA⌕ Search

SEARCH · Search NASA

Results for “Pattern 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 109 records · Page 6
All sourcesNASA NTRSDOE OSTI
NASA NTRS1986

Microcomputer-based programmable optical signal processor

A microcomputer-based real-time programmable optical signal processing system utilizing a magneto-optic spatial light modulator (MOSLM) and a liquid crystal light valve (LCLV) is described. This system can perform a myriad of complicated optical operations, such as image correlation, image subtraction, and matrix multiplication. Its important asserts are its programmability and the capability of real-time addressing. These are important for telerobotic vision in space automation applications. Design specifications and suggestions for practical implementation of the system are discussed. Some preliminary experimental demonstrations are conducted to demonstrate applications of the proposed system to image correlation for optical pattern recognition, image subtraction for lC chip inspection, and matrix multiplication for optical computing.

Yu, F. T. S.↗
DOE OSTI2025

How does ion temperature gradient turbulence depend on magnetic geometry? Insights from data and machine learning

Magnetic geometry has a significant effect on the level of turbulent transport in fusion plasmas. Here, we model and analyse this dependence using multiple machine learning methods and a dataset of >200 000 nonlinear gyrokinetic simulations of ion-temperature-gradient turbulence in diverse non-axisymmetric geometries. The dataset is generated using a large collection of both optimised and randomly generated stellarator equilibria. At fixed gradients and other input parameters, the turbulent heat flux varies between geometries by several orders of magnitude. Trends are apparent among the configurations with particularly high or particularly low heat flux. Regression and classification techniques from machine learning are then applied to extract patterns in the dataset. Due to a symmetry of the gyrokinetic equation, the heat flux and regressions thereof should be invariant to translations of the raw features in the parallel coordinate, similar to translation invariance in computer vision applications. Multiple regression models including convolutional neural networks (CNNs) and decision trees can achieve reasonable predictive power for the heat flux in held-out test configurations, with highest accuracy for the CNNs. Using Spearman correlation, sequential feature selection and Shapley values to measure feature importance, it is consistently found that the most important geometric lever on the heat flux is the flux surface compression in regions of bad curvature. The second most important geometric feature relates to the magnitude of geodesic curvature. These two features align remarkably with surrogates that have been proposed based on theory, while the methods here allow a natural extension to more features for increased accuracy. The dataset, released with this publication, may also be used to test other proposed surrogates, and we find that many previously published proxies do correlate well with both the heat flux and stability boundary.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
NASA NTRS1981

Laser electro-optic system for rapid three-dimensional /3-D/ topographic mapping of surfaces

It is pointed out that the generic utility of a robot in a factory/assembly environment could be substantially enhanced by providing a vision capability to the robot. A standard videocamera for robot vision provides a two-dimensional image which contains insufficient information for a detailed three-dimensional reconstruction of an object. Approaches which supply the additional information needed for the three-dimensional mapping of objects with complex surface shapes are briefly considered and a description is presented of a laser-based system which can provide three-dimensional vision to a robot. The system consists of a laser beam array generator, an optical image recorder, and software for controlling the required operations. The projection of a laser beam array onto a surface produces a dot pattern image which is viewed from one or more suitable perspectives. Attention is given to the mathematical method employed, the space coding technique, the approaches used for obtaining the transformation parameters, the optics for laser beam array generation, the hardware for beam array coding, and aspects of image acquisition.

Altschuler, M. D.↗
NASA NTRS1996

Development of Collaborative Research Initiatives to Advance the Aerospace Sciences-via the Communications, Electronics, Information Systems Focus Group

The primary goal of the Adaptive Vision Laboratory Research project was to develop advanced computer vision systems for automatic target recognition. The approach used in this effort combined several machine learning paradigms including evolutionary learning algorithms, neural networks, and adaptive clustering techniques to develop the E-MOR.PH system. This system is capable of generating pattern recognition systems to solve a wide variety of complex recognition tasks. A series of simulation experiments were conducted using E-MORPH to solve problems in OCR, military target recognition, industrial inspection, and medical image analysis. The bulk of the funds provided through this grant were used to purchase computer hardware and software to support these computationally intensive simulations. The payoff from this effort is the reduced need for human involvement in the design and implementation of recognition systems. We have shown that the techniques used in E-MORPH are generic and readily transition to other problem domains. Specifically, E-MORPH is multi-phase evolutionary leaming system that evolves cooperative sets of features detectors and combines their response using an adaptive classifier to form a complete pattern recognition system. The system can operate on binary or grayscale images. In our most recent experiments, we used multi-resolution images that are formed by applying a Gabor wavelet transform to a set of grayscale input images. To begin the leaming process, candidate chips are extracted from the multi-resolution images to form a training set and a test set. A population of detector sets is randomly initialized to start the evolutionary process. Using a combination of evolutionary programming and genetic algorithms, the feature detectors are enhanced to solve a recognition problem. The design of E-MORPH and recognition results for a complex problem in medical image analysis are described at the end of this report. The specific task involves the identification of vertebrae in x-ray images of human spinal columns. This problem is extremely challenging because the individual vertebra exhibit variation in shape, scale, orientation, and contrast. E-MORPH generated several accurate recognition systems to solve this task. This dual use of this ATR technology clearly demonstrates the flexibility and power of our approach.

Knasel, T. Michael↗
DOE OSTI2025

RingX: Scalable Parallel Attention for Long-Context Learning on HPC

The attention mechanism has become foundational for remarkable AI breakthroughs since the introduction of the Transformer, driving the demand for increasingly longer context to power frontier models such as large-scale reasoning language models and high-resolution image/video generators. However, its quadratic computational and memory complexities present substantial challenges. Current state-of-the-art parallel attention methods, such as ring attention, are widely adopted for long-context training but utilize a point-to-point communication strategy that fails to fully exploit the capabilities of modern HPC network architectures. In this work, we propose ringX, a scalable family of parallel attention methods optimized explicitly for HPC systems. By enhancing workload partitioning, refining communication patterns, and improving load balancing, ringX achieves up to 3.4 × speedup compared to conventional ring attention on the Frontier supercomputer. Optimized for both bi-directional and causal attention mechanisms, ringX demonstrates its effectiveness through training benchmarks of a Vision Transformer (ViT) on a climate dataset and a Generative Pre-Trained Transformer (GPT) model, Llama3 8B. Our method attains an end-to-end training speedup of approximately 1.5 × in both scenarios. To our knowledge, the achieved 38% model FLOPs utilization (MFU) for training Llama3 8B with a 1M-token sequence length on 4,096 GPUs represents one of the highest training efficiencies reported for long-context learning on HPC systems. Our code implementation is available at https://github.com/jqyin/ringX-attention.

Yin, Junqi [ORNL] (ORCID:0000000338435520)↗
NASA NTRS2014

Vision Algorithm for the Solar Aspect System of the High Energy Replicated Optics to Explore the Sun Mission

This work covers the design and test of a machine vision algorithm for generating high- accuracy pitch and yaw pointing solutions relative to the sun on a high altitude balloon. It describes how images were constructed by focusing an image of the sun onto a plate printed with a pattern of small cross-shaped fiducial markers. Images of this plate taken with an off-the-shelf camera were processed to determine relative position of the balloon payload to the sun. The algorithm is broken into four problems: circle detection, fiducial detection, fiducial identification, and image registration. Circle detection is handled by an "Average Intersection" method, fiducial detection by a matched filter approach, and identification with an ad-hoc method based on the spacing between fiducials. Performance is verified on real test data where possible, but otherwise uses artificially generated data. Pointing knowledge is ultimately verified to meet the 20 arcsecond requirement.

0000000↗
NASA NTRSDate not supplied

3D Localization of Defects in Facility Inspection

Wind tunnels are crucial facilities that support the aerospace industry. However, these facilities are large, complex, and pose unique maintenance and inspection requirements. Manual inspections to identify defects such as cracks, missing fasteners, leaks, and foreign objects are important but labor and schedule intensive. Our goal is to utilize small Unmanned Aircraft Systems (sUAS) and computer vision-based analysis to automate the inspection of the interior and exterior of NASA’s critical wind tunnel facilities. We detect missing fasteners as our defect class, and detect existing fasteners to provide potential future missing fastener sites for preventative maintenance. These detections are done on both 2D raw images and in 3D space to provide a visual reference and real world location to facilitate repairs. A dataset was created consisting of images taken along a grid-like pattern of an interior tunnel section in the AEDC National Full-Scale Aerodynamics Complex (NFAC) at NASA Ames Research Center. Our method uses object detection to create image level bounding boxes of the fasteners and missing fasteners, then uses photogrammetry to create a mapping from 2D image locations to 3D real world locations. The image level bounding boxes and the 2D to 3D mapping are then combined to determine the 3D location of the defects. We describe the data collection, photogrammetry, and computer vision techniques used for object detection as well as a quantitative analysis of the method.

Small Unmanned Aircraft Systems (sUAS)↗
NASA NTRS2007

Stroboscopic Vision as a Treatment for Retinal Slip Induced Motion Sickness

Motion sickness in the general population is a significant problem driven by the increasingly more sophisticated modes of transportation, visual displays, and virtual reality environments. It is important to investigate non-pharmacological alternatives for the prevention of motion sickness for individuals who cannot tolerate the available anti-motion sickness drugs, or who are precluded from medication because of different operational environments. Based on the initial work of Melvill Jones, in which post hoc results indicated that motion sickness symptoms were prevented during visual reversal testing when stroboscopic vision was used to prevent retinal slip, we have evaluated stroboscopic vision as a method of preventing motion sickness in a number of different environments. Specifically, we have undertaken a five part study that was designed to investigate the effect of stroboscopic vision (either with a strobe light or LCD shutter glasses) on motion sickness while: (1) using visual field reversal, (2) reading while riding in a car (with or without external vision present), (3) making large pitch head movements during parabolic flight, (4) during exposure to rough seas in a small boat, and (5) seated and reading in the cabin area of a UH60 Black Hawk Helicopter during 20 min of provocative flight patterns.

Reschke, M. F.↗
DOE OSTI2026

Selective Amnesia using Contrastive Subnet Erasure for Class Level Unlearning in Vision Models

We study concept-level forgetting in pretrained vision models: removing an entire semantic category so the system no longer recognizes that object in unseen images and contexts, rather than merely forgetting specific training examples. Prior work either applies blunt global projections or fine-tunes parameters, which can introduce collateral damage to unrelated features, add compute, and become unstable as forgetting strength increases. We introduce Contrastive Subnet Erasure (CSE), a training-free, encoder-centric edit that targets a compact set of channels most responsible for the class and attenuates them in a calibrated manner. The modification is algebraically folded into the subsequent layer, yielding no inference-time overhead and leaving task heads unchanged. To evaluate whether forgetting generalizes beyond the data used to specify the class, we introduce a cross dataset protocol in which the class is defined on a source dataset and performance is measured on a disjoint target dataset drawn from a different distribution with no shared images. This setup tests whether the model still fails to recognize the object when it looks different or appears in new scenes, and it helps avoid overfitting to patterns in the source dataset. Across CIFAR 10, CIFAR 100, and ImageNet under this protocol, CSE achieves stronger forgetting of the target class while better preserving non target utility than existing baselines in both single class and multi class settings. Overall, CSE provides a simple, stable, and deployment-ready mechanism for class-level unlearning in vision.

Kotevska, Olivera [ORNL] (ORCID:0000000316772243)↗
NASA NTRS2020

Evaluation of Tablet-Based Methods for Assessment of Contrast Sensitivity

Some astronauts have suffered degradation of vision during long-duration space flight, suffering from a condition that has come to be known as Spaceflight Associated Neuro-ocular Syndrome (SANS). While related morphological changes can be observed with imaging technologies such as optical coherence tomography (OCT), it may be useful to have a rapid method for functional vision assessment. In this paper, we compare three tablet-based methods for rapid assessment of contrast sensitivity. First, a relatively novel method developed expressly for touch screens, in which the subject ”swipes” a frequency/contrast sweep grating to indicate the boundary between visible and invisible patterns; second, a method-of-adjustment task in which the subject adjusts the contrast of a grating patch up and down to bracket the visual threshold; and third, a traditional temporal two-alternative forced choice (2AFC) task, in which the subject is presented with a near-threshold stimulus in one of two intervals, and must report the interval containing the stimulus. The swipe method shows variability comparable to the 2AFC method, and shows good agreement in estimates of the spatial frequency of peak sensitivity. The absolute sensitivity estimated with the swipe method is higher than that of the other methods, perhaps because subjects are biased to trace outside of the visible pattern region, or perhaps due to stimulus differences.

Mulligan, Jeffrey B.↗
NASA NTRS2015

Identification and Evaluation of Integration and Cross Cutting Issues Across HRP Risks

The HRP Integrated Research Plan contains the research plans for the 32 risks requiring research to characterize and mitigate. These risks to human health and performance in spaceflight are identified by evidence and each one focuses on a single aspect of human physiology or performance. They are further categorized by aspects of the spaceflight environment, such as altered gravity or space radiation, that that play a major role in their likelihood and consequence. From its inception the "integrate" in the Research Plan has denoted the integrated nature of risks to human health and performance, the connectedness of physiological systems within the human body regardless of the spaceflight environment, and the integrated response of the human body to the spaceflight environment. Common characteristics of the spaceflight environment include altered gravity, atmospheres and light/dark cycles, space radiation, isolation, noise, and periods of high or low workload. Long term exposure to this unique environment produces a suite of physiological effects such as stress; vision, neurocognitive and anthropometric changes; circadian misalignment; fluid shifts, deconditioning; immune dysregulation; and altered nutritional requirements. Matrix diagraming was used to systematically identify, analyze and rate the many-to-many relationships between environmental characteristics and the suite of physiological effects. It was also to identify patterns in the relationships of common physiological effects to each other. Analyses of patterns or relationships in these diagrams help to identify issues that cut across multiple risks. Cross cutting issues benefit from a multidisciplinary approach that synthesizes concepts or data from two or more disciplines to identify and characterize risk factors or develop countermeasures relevant to multiple risks. They also help to illuminate possible problem areas that may arise when a countermeasure impacts risks other than those which it was developed to mitigate, or identify groupings of physiological changes that are likely to occur that may impact the overall risk posture.

Steinberg, S. L.↗
NASA NTRS1997

Digital Images and Human Vision

Processing of digital images destined for visual consumption raises many interesting questions regarding human visual sensitivity. This talk will survey some of these questions, including some that have been answered and some that have not. There will be an emphasis upon visual masking, and a distinction will be drawn between masking due to contrast gain control processes, and due to processes such as hypothesis testing, pattern recognition, and visual search.

Watson, Andrew B.↗
NASA NTRS2014

Vision Algorithm for the Solar Aspect System of the HEROES Mission

This work covers the design and test of a machine vision algorithm for generating high-accuracy pitch and yaw pointing solutions relative to the sun for the High Energy Replicated Optics to Explore the Sun (HEROES) mission. It describes how images were constructed by focusing an image of the sun onto a plate printed with a pattern of small fiducial markers. Images of this plate were processed in real time to determine relative position of the balloon payload to the sun. The algorithm is broken into four problems: circle detection, fiducial detection, fiducial identification, and image registration. Circle detection is handled by an "Average Intersection" method, fiducial detection by a matched filter approach, identification with an ad-hoc method based on the spacing between fiducials, and image registration with a simple least squares fit. Performance is verified on a combination of artificially generated images, test data recorded on the ground, and images from the 2013 flight

Solar Aspect↗
NASA NTRS2014

Vision Algorithm for the Solar Aspect System of the HEROES Mission

This work covers the design and test of a machine vision algorithm for generating high-accuracy pitch and yaw pointing solutions relative to the sun for the High Energy Replicated Optics to Explore the Sun (HEROES) mission. It describes how images were constructed by focusing an image of the sun onto a plate printed with a pattern of small fiducial markers. Images of this plate were processed in real time to determine relative position of the balloon payload to the sun. The algorithm is broken into four problems: circle detection, fiducial detection, fiducial identification, and image registration. Circle detection is handled by an Average Intersection method, fiducial detection by a matched filter approach, identification with an ad-hoc method based on the spacing between fiducials, and image registration with a simple least squares fit. Performance is verified on a combination of artificially generated images, test data recorded on the ground, and images from the 2013 flight.

HEROES↗
NASA NTRSDate not supplied

Ground to Air Testing of a Fused Optical-Radar Aircraft Detection and Tracking System

Onboard detection and tracking capability are integral to the sensing component in collision avoidance systems needed to safely operate autonomous urban air taxis and small Unmanned Aircraft Systems. Ground-based validation of detection and tracking systems is an important milestone towards the end goal of real-time collision avoidance using onboard sensors and algorithms. In this work, we evaluate three Extended Kalman Filter(EKF)based fusion trackers with radar and vision detection inputs, and compare them with baseline trackers for each sensor type. Performance is assessed using field collected data of ground to air test flights with the sensors co-located on a stable platform with an instrumented multirotor acting as the intruder performing a waypoint pattern at a distance of1.1km to 0.3km to simulate a head-on collision geometry. Fusing an image-based morphological detector with a radar detector using an EKF covered 74% of the ground truth position updates logged by the flight controller on the multirotor within 50 meters after accounting for alignment offsets while covering 15% more ground truth updates relative to radar only. Removing timestamps when the intruder aircraft is occluded by trees and only considering timestamps where the radar has an update, the EKF image-based morphological detector combined with the radar detector covered 90% of the of the ground truth position updates within 50metersand 97% within 100 meters.

Chester V Dolph↗
NASA NTRS1989

Optical pattern recognition; Proceedings of the Meeting, Los Angeles, CA, Jan. 17, 18, 1989

Papers on optical pattern recognition are presented, covering topics such as the estimation of satellite pose and motion parameters using a neural net tracker, associative memory, optical implmentation of programmable neural networks, optoelectronic neural networks, dynamic autoassociative neural memory, heteroassociative memory, bilinear pattern recognition processors, optical processing of optical correlation plane data, and a synthetic discriminant function-based nonlinear optical correlator. Other topics include an interactive optical-digital image processor, geometric transformations for video compression and human teleoperator display, quasiconformal remapping for compensation of human visual field defects, hybrid vision for automated spacecraft landing, advanced symbolic and inference optical correlation filters, and a rotationally invariant holographic tracking system. Additional topics include the detection of rotational and scale-varying objects with a programmable joint transform correlator, a single spatial light modulator binary nonlinear optical correlator, optical joint transform correlation, linear phase coefficient composite filters, and binary phase-only filters.

Liu, Hua-Kuang↗
NASA NTRS1990

Image recovery from edge primitives

A method for extracting edge primitives from Mach-band patterns is presented together with a method for recovering image representations of features outlined by the edge boundaries. The accuracy, stability, and resolution of these representations are assessed. Since these representations are most commonly used in characterizing targets, this method of low-level processing offers new opportunities for computer vision and high data-compressing coding. Two bandpass filters are considered, the spatially invariant Laplacian of Gaussian filter and spatially variant intensity-dependent spatial (IDS) summation. It is shown that the recovery from the IDS bandpass data is particularly advantageous in applications for which robustness to local and temporal variations in illumination is important. It is concluded that the edge primitives extracted from bandpassed images can be an efficient way to store, transmit, and represent images.

Alter-Gartenberg, Rachel↗
NASA NTRS1992

Space flight and changes in spatial orientation

From a sensory point of view, space flight represents a form of stimulus rearrangement requiring modification of established terrestrial response patterns through central reinterpretation. Evidence of sensory reinterpretation is manifested as postflight modifications of eye/head coordination, locomotor patterns, postural control strategies, and illusory perceptions of self or surround motion in conjunction with head movements. Under normal preflight conditions, the head is stabilized during locomotion, but immediately postflight reduced head stability, coupled with inappropriate eye/head coordination, results in modifications of gait. Postflight postural control exhibits increased dependence on vision which compensates for inappropriate interpretation of otolith and proprioceptive inputs. Eye movements compensatory for perceived self motion, rather than actual head movements have been observed postflight. Overall, the in-flight adaptive modification of head stabilization strategies, changes in head/eye coordination, illusionary motion, and postural control are maladaptive for a return to the terrestrial environment. Appropriate countermeasures for long-duration flights will rely on preflight adaptation and in-flight training.

Reschke, Millard F.↗