Reconfigurable Sensor Fusion Processing Hardware
The paper presents the concept and initial test from the hardware implementation of a low-power, high-speed reconfigurable sensor fusion processor.
Hardware implementation multisensor fusion data fu↗
SEARCH · Engineering Papers
Results for “image recognition”
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.
Knowledge Base for Distributed Spacecraft Mission Design Using the Trade-Space Analysis Tool for Constellations (TAT-C)
Opportunities for multi-point measurements, greater revisit frequency, failure robustness, and improved cost effectiveness motivate consideration of Distributed Spacecraft Missions (DSMs) for future Earth science missions. However, careful analysis is required to assess the distributed sensing capabilities of a constellation compared to more mature monolithic spacecraft while also considering other important dimensions such as cost and risk. The large combinatorial DSM design space limits existing mission analysis tools and exploration methods which emphasize monolithic design variables. The Trade-space Analysis Tool for Constellations (TAT-C) under development at Goddard seeks to enumerate and evaluate alternative mission architectures to minimize cost and maximize scientific return for pre-defined goals during pre-phase A analysis.Similar to other model-centric engineering efforts, efficient data management is a significant challenge for DSM mission analysis. In TAT-C, a Knowledge Base (KB) is envisioned as a cumulative central repository of information and meta-information about DSMs. Initial KB concepts store related data for reuse within or across mission analyses; however, over time, the KB is envisioned to be an important layer to coordinate actions of both human analysts and automated design agents to search a large design space for desirable mission alternatives. Preliminary KB research builds on a modern web technology stack to provide the following functionality: 1) storage of trade-space search requests which set requirements and constraints for DSM concepts, 2) storage of analysis results which quantify performance metrics for evaluated DSM concepts, 3) a RESTful application programming interface (API) for scripted access to data from TAT-C modules, 4) web-based graphical user interface (GUI) for manual access to underlying data, and 5) access control and management restrictions relevant to data protection and security. These efforts have culminated in a prototype KB used by the research team during TAT-C development to assess opportunities for future work.
Pattern Recognition↗
Taming the Untamable: Making An Enceladus-Like Plume by Hypervelocity Oblique Impacts at the NASA Ames Vertical Gun Range (AVGR) Facility
Enceladus’ plume was extensively investigated by the Cassini-Huygens mission [1-4], giving information about the icy moon interior, and the existence of a plausibly inhabited global ocean. The plume is a mixture of gas, sub-micrometer to micrometer ice, and non-ice particles, expanding into space from vents in the Tiger Stripes region [5]. Collecting plume’s ice particles for signs of Life is a fundamental objective of future fly-by missions to Enceladus, and other icy moons, e.g., Europa, where icy plumes could exist. However, understanding particle collection on high-speed flybys through a plume environment is a challenging task. To address the above, in collaboration with the Applied Physics Laboratory (APL), we have been testing the collection of impact-generated ice grains from simulated ice plumes at the NASA Ames Vertical GunRange (AVGR) facility [e.g., 6-7].Of particular interest is the alteration of organic material in the particles during the high-speed collection process. Any such alteration could degrade the ability of instruments to detect biomarkers in the samples collected. An experimental and theoretical understanding of the organic alteration as a function of collection speed could be used to unravel these effects. Methods: Plumes of -170°C ice particles traveling at an initial speed of 2 – 3 km/s were created by shooting 3-mm hollow Aluminum (Al) projectiles (up to 6 km/s) at liquid-nitrogen-cooled ice targets of synthetic seawater (SSW). An Al witness foil is positioned to determine particle size, and ice particle impact speed is determined by high-speed video. The ice particles smash and sublimate on impact leaving salt residue, dents, and pits on the Al witness foil. We characterized Ice grains type, abundance, density, shape, and distribution caliper (Ferret)diameter via image pattern recognition analysis (ImageJ Software) of still camera photographs of the Al target. Results: Our hyper-velocity impact experiments generated an Enceladus plume analog with an ice particle mean diameter within the particle size distribution (i.e., 2 micrometers to 50 micrometers) as determined by Cassini’s multiple instruments [8]. Simulated impact speed ranged0.2 km/s to 1-2 km/s. The continued experimental production of high-velocity, micrometer-sized ice grains is underway at the AVGR. The relevance of ice particle generation is multifold, so are its applications. These include testing for fragmentation of molecular organics to interpret Cassini’s data (e.g., Cassini Data Analysis Program) and studies of forwarding contamination (a planetary protection issue).
Rosalba Bonaccorsi↗
Artificial Intelligence Application to D and D - 20492
As aging facilities across the DOE complex await decommissioning, there is an ongoing need to understand any changes in the structural conditions. Many of these facilities were built over 50 years ago and, in some cases, these facilities have gone beyond the expected operational lifetime. Many facilities have been placed in a state of 'cold and dark,' sitting unused and awaiting decommissioning. Especially challenging are the aging facilities that provide unique operational/production capabilities to support critical DOE missions and cannot be shut down. In any of these scenarios, the structural integrity of these facilities may become compromised as time passes. It is critical that adequate inspections be performed on a continual basis and that the data collected undergoes sufficient analysis to support timely identification of any new or worsening structural issues as well as prompt needed maintenance and repairs to maintain the facilities in a safe condition. In recent days, Artificial Intelligence (AI) [1] and its application to various domains are growing at fast speed. FIU is performing research in this area and exploring the associated technologies to solve nuclear decommissioning problems. Artificial intelligence refers to the capability of a program to autonomously act, react and adapt to the working environment. AI enables the machine to behave like humans and perform the cognitive functions such as 'learning' and 'problem solving'. AI systems gradually moving from traditional approaches (algorithms and expert systems) towards more efficient and advanced technologies (machine learning [1] and deep learning [2] [3]). AI is the study of algorithms and statistical models that is being used by computers to perform specific tasks without using explicit instructions. FIU is working to develop a pilot-scale infrastructure to implement structural health monitoring using AI technologies with focus on machine learning, deep learning. This research is focused on Computer Vision/Image Classification area of AI applications. This can also be expanded to other areas of AI related to Object Recognition and Character Recognition in images. In addition to utilizing existing data sets, FIU will collect and investigate image and video data using FIU test-bed mockups to monitor structural health of the facility. Resulting data will be processed and analyzed using machine learning/deep learning technologies. The proposed pilot system is intended to serve as a starting point to engage the DOE field sites on related data sets and their decision making needs. It is anticipated that proposed machine learning/deep learning technologies can be effectively employed using anomaly detection to solve EM challenges in surveillance and maintenance of the D and D facilities. FIU will work with research stakeholders to identify applications at various sites and other DOE facilities. (authors)
12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
Fault tectonics and earthquake hazards in parts of southern California
The author has identified the following significant results. Four previously unknown faults were discovered in basement terrane of the Peninsular Ranges. These have been named the San Ysidro Creek fault, Thing Valley fault, Canyon City fault, and Warren Canyon fault. In addition fault gouge and breccia were recognized along the San Diego River fault. Study of features on Skylab imagery and review of geologic and seismic data suggest that the risk of a damaging earthquake is greater along the northwestern portion of the Elsinore fault than along the southeastern portion. Physiographic indicators of active faulting along the Garlock fault identifiable in Skylab imagery include scarps, linear ridges, shutter ridges, faceted ridges, linear valleys, undrained depressions and offset drainage. The following previously unrecognized fault segments are postulated for the Salton Trough Area: (1) An extension of a previously known fault in the San Andreas fault set located southeast of the Salton Sea; (2) An extension of the active San Jacinto fault zone along a tonal change in cultivated fields across Mexicali Valley ( the tonal change may represent different soil conditions along opposite sides of a fault). For the Skylab and LANDSAT images studied, pseudocolor transformations offer no advantages over the original images in the recognition of faults in Skylab and LANDSAT images. Alluvial deposits of different ages, a marble unit and iron oxide gossans of the Mojave Mining District are more readily differentiated on images prepared from ratios of individual bands of the S-192 multispectral scanner data. The San Andreas fault was also made more distinct in the 8/2 and 9/2 band ratios by enhancement of vegetation differences on opposite sides of the fault. Preliminary analysis indicates a significant earth resources potential for the discrimination of soil and rock types, including mineral alteration zones. This application should be actively pursued.
Merifield, P. M.↗
Recognition of 3-D symmetric objects from range images in automated assembly tasks
A new technique is presented for the three dimensional recognition of symmetric objects from range images. Beginning from the implicit representation of quadrics, a set of ten coefficients is determined for symmetric objects like spheres, cones, cylinders, ellipsoids, and parallelepipeds. Instead of using these ten coefficients trying to fit them to smooth surfaces (patches) based on the traditional way of determining curvatures, a new approach based on two dimensional geometry is used. For each symmetric object, a unique set of two dimensional curves is obtained from the various angles at which the object is intersected with a plane. Using the same ten coefficients obtained earlier and based on the discriminant method, each of these curves is classified as a parabola, circle, ellipse, or hyperbola. Each symmetric object is found to possess a unique set of these two dimensional curves whereby it can be differentiated from the others. It is shown that instead of using the three dimensional discriminant which involves evaluation of the rank of its matrix, it is sufficient to use the two dimensional discriminant which only requires three arithmetic operations.
Alvertos, Nicolas↗
Versatile recognition of graphene layers from optical images under controlled illumination through green channel correlation method
In this study, a simple yet versatile method is proposed for identifying the number of exfoliated graphene layers transferred on an oxide substrate from optical images, utilizing a limited number of input images for training, paired with a more traditional number of a few thousand well-published Github images for testing and predicting. Two thresholding approaches, namely the standard deviation-based approach and the linear regression-based approach, were employed in this study. The method specifically leverages the red, green, and blue color channels of image pixels and creates a correlation between the green channel of the background and the green channel of the various layers of graphene. This method proves to be a feasible alternative to deep learning-based graphene recognition and traditional microscopic analysis. The proposed methodology performs well under conditions where the effect of surrounding light on the graphene-on-oxide sample is minimum and allows rapid identification of the various graphene layers. Here, the study additionally addresses the functionality of the proposed methodology with nonhomogeneous lighting conditions, showcasing successful prediction of graphene layers from images that are lower in quality compared to typically published in literature. In all, the proposed methodology opens up the possibility for the non-destructive identification of graphene layers from optical images by utilizing a new and versatile method that is quick, inexpensive, and works well with fewer images that are not necessarily of high quality.
36 MATERIALS SCIENCE↗
Studying the Sky/Planets Can Drown You in Images: Machine Learning Solutions at JPL/Caltech
JPL is working to develop a domain-independent system capable of small-scale object recognition in large image databases for science analysis. Two applications discussed are the cataloging of three billion sky objects in the Sky Image Cataloging and Analysis Tool (SKICAT) and the detection of possibly one million small volcanoes visible in the Magellan synthetic aperture radar images of Venus (JPL Adaptive Recognition Tool, JARTool).
geomorphology↗
Discriminating coastal rangeland production and improvements with computer aided techniques
The feasibility and utility of using satellite data and computer-aided remote sensing analysis techniques to conduct range inventories were tested. This pilot study was focused over a 250,000 acre site in Galveston and Brazoria Counties along the Texas Gulf Coast. Rectified enlarged aircraft color infrared photographs of this site were used as the ground truth base. The different land categories were identified, delineated, and measured. Multispectral scanner (MSS) bulk data from LANDSAT-1 was received and analyzed with the Image 100 pattern recognition system. Features of interest were delineated on the image console giving the number of picture elements classified; the picture elements were converted to acreages and the accuracy of the technique was evaluated by comparison with data base results for three test sites. The accuracies for computer aided classification of coastal marshes ranged from 89% to 96%.
Reeves, C. A.↗
An optical processor for object recognition and tracking
The design and development of a miniaturized optical processor that performs real time image correlation are described. The optical correlator utilizes the Vander Lugt matched spatial filter technique. The correlation output, a focused beam of light, is imaged onto a CMOS photodetector array. In addition to performing target recognition, the device also tracks the target. The hardware, composed of optical and electro-optical components, occupies only 590 cu cm of volume. A complete correlator system would also include an input imaging lens. This optical processing system is compact, rugged, requires only 3.5 watts of operating power, and weighs less than 3 kg. It represents a major achievement in miniaturizing optical processors. When considered as a special-purpose processing unit, it is an attractive alternative to conventional digital image recognition processing. It is conceivable that the combined technology of both optical and ditital processing could result in a very advanced robot vision system.
Sloan, J.↗
Advanced Video Guidance Sensor and Next Generation Autonomous Docking Sensors
In recent decades, NASA's interest in spacecraft rendezvous and proximity operations has grown. Additional instrumentation is needed to improve manned docking operations' safety, as well as to enable telerobotic operation of spacecraft or completely autonomous rendezvous and docking. To address this need, Advanced Optical Systems, Inc., Orbital Sciences Corporation, and Marshall Space Flight Center have developed the Advanced Video Guidance Sensor (AVGS) under the auspices of the Demonstration of Autonomous Rendezvous Technology (DART) program. Given a cooperative target comprising several retro-reflectors, AVGS provides six-degree-of-freedom information at ranges of up to 300 meters for the DART target. It does so by imaging the target, then performing pattern recognition on the resulting image. Longer range operation is possible through different target geometries. Now that AVGS is being readied for its test flight in 2004, the question is: what next? Modifications can be made to AVGS, including different pattern recognition algorithms and changes to the retro-reflector targets, to make it more robust and accurate. AVGS could be coupled with other space-qualified sensors, such as a laser range-and-bearing finder, that would operate at longer ranges. Different target configurations, including the use of active targets, could result in significant miniaturization over the current AVGS package. We will discuss these and other possibilities for a next-generation docking sensor or sensor suite that involve AVGS.
Granade, Stephen R.↗
Automatic melt pool recognition in X-ray radiography images from laser-molten Al alloy
Not Available
36 MATERIALS SCIENCE↗
Recursive, in-place algorithm for the hexagonal orthogonal oriented quadrature image pyramid
Pyramid image transforms have proven useful in image coding and pattern recognition. The hexagonal orthogonal oriented quadrature image pyramid (HOP), transforms an image into a set of orthogonal, oriented, odd and even bandpass subimages. It operates on a hexagonal input lattice and employs seven kernels, each of which occupies a neighborhood consisting of a point and a hexagon of six nearest neighbors. The kernels consist of one lowpass and six bandpass kernels that are orthogonal, self-similar, and localized in space, spatial frequency, orientation, and phase. The kernels are first applied to the image samples to create the first level of the pyramid, then to the lowpass coefficients to create the next level. The resulting pyramid is a compact, efficient image code. Here, a recursive, in-place algorithm for computation of the HOP transform is described. The transform may be regarded as a depth-first traversal of a tree structure. It is shown that the algorithm requires a number of operations that is on the order of the number of pixels.
Watson, Andrew B.↗
Digital mammography, cancer screening: Factors important for image compression
The use of digital mammography for breast cancer screening poses several novel problems such as development of digital sensors, computer assisted diagnosis (CAD) methods for image noise suppression, enhancement, and pattern recognition, compression algorithms for image storage, transmission, and remote diagnosis. X-ray digital mammography using novel direct digital detection schemes or film digitizers results in large data sets and, therefore, image compression methods will play a significant role in the image processing and analysis by CAD techniques. In view of the extensive compression required, the relative merit of 'virtually lossless' versus lossy methods should be determined. A brief overview is presented here of the developments of digital sensors, CAD, and compression methods currently proposed and tested for mammography. The objective of the NCI/NASA Working Group on Digital Mammography is to stimulate the interest of the image processing and compression scientific community for this medical application and identify possible dual use technologies within the NASA centers.
Clarke, Laurence P.↗
Automatic threat recognition system and method using material disambiguation informed by physics in x-ray CT images of baggage
An automatic threat recognition (ATR) system is disclosed for scanning an article to recognize contraband items or items of interest contained within the article. The ATR system uses a CAT scanner to obtain a CT image scan of objects within the article, representing a plurality of 2D image slices of the article and its contents. Each 2D image slice includes information forming a plurality of voxels. The ATR system includes a computer and determines which voxels have a likelihood of representing materials of interest. It then aggregates those voxels to produce detected objects. The detected objects are further classified as items of interest vs. not of interest. The ATR system is based on learned parameters for a novel interaction of global and object context mechanisms. ATR system performance may be optimized by using jointly optimal global and object context parameters learned during training. The global context parameters may apply to the article as a whole and facilitate object detection. The object context parameters may apply to the individual object detections.
Paglieroni, David W.↗
Computer processed LANDSAT data
Background information and exercises are provided to: (1) establish or expand understanding of the concepts, methods, and terminology of computer processing of image producing data; (2) develop insight into the advantages of computer based image processing compared with the photointerpretation approach for processing, classifying, interpreting, and applying remote sensing data; (3) foster a broad perspective on the principal of the main techniques for image enhancement, pattern recognition, and thematic classification; (4) appreciate the pros and cons of batch and interactive modes of image analysis; (5) examine and evaluate some specific computer generated products for subscenes in Pennsylvania and New Jersey; and (6) interrelate these particular examples of output with more theoretical explanations of computer processing strategies and procedures.
Source record↗
Computer vision research at Marshall Space Flight Center
Orbital docking, inspection, and sevicing are operations which have the potential for capability enhancement as well as cost reduction for space operations by the application of computer vision technology. Research at MSFC has been a natural outgrowth of orbital docking simulations for remote manually controlled vehicles such as the Teleoperator Retrieval System and the Orbital Maneuvering Vehicle (OMV). Baseline design of the OMV dictates teleoperator control from a ground station. This necessitates a high data-rate communication network and results in several seconds of time delay. Operational costs and vehicle control difficulties could be alleviated by an autonomous or semi-autonomous control system onboard the OMV which would be based on a computer vision system having capability to recognize video images in real time. A concept under development at MSFC with these attributes is based on syntactic pattern recognition. It uses tree graphs for rapid recognition of binary images of known orbiting target vehicles. This technique and others being investigated at MSFC will be evaluated in realistic conditions by the use of MSFC orbital docking simulators. Computer vision is also being applied at MSFC as part of the supporting development for Work Package One of Space Station Freedom.
Vinz, Frank L.↗
Some methods of encoding simple visual images for use with a sparse distributed memory, with applications to character recognition
To study the problems of encoding visual images for use with a Sparse Distributed Memory (SDM), I consider a specific class of images- those that consist of several pieces, each of which is a line segment or an arc of a circle. This class includes line drawings of characters such as letters of the alphabet. I give a method of representing a segment of an arc by five numbers in a continuous way; that is, similar arcs have similar representations. I also give methods for encoding these numbers as bit strings in an approximately continuous way. The set of possible segments and arcs may be viewed as a five-dimensional manifold M, whose structure is like a Mobious strip. An image, considered to be an unordered set of segments and arcs, is therefore represented by a set of points in M - one for each piece. I then discuss the problem of constructing a preprocessor to find the segments and arcs in these images, although a preprocessor has not been developed. I also describe a possible extension of the representation.
Jaeckel, Louis A.↗