Engineering Papers⌕ Search

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.

At least 505 records · Page 28
All sourcesNASA NTRSDOE OSTI
NASA NTRS1974

Management of natural resources through automatic cartographic inventory

The author has identified the following significant results. (1) Accurate recognition of previously known ground features from ERTS-1 imagery has been confirmed and a probable detection range for the major signatures can be given. (2) Unidentified elements, however, must be decoded by means of the equal densitometric value zone method. (3) Determination of these zonings involves an analogical treatment of images using the color equidensity methods (pseudo-color), color composites and especially temporal color composite (repetitive superposition). (4) After this analogical preparation, the digital equidensities can be processed by computer in the four MSS bands, according to a series of transfer operations from imagery and automatic cartography.

Rey, P.↗
NASA NTRS2012

Localization Using Visual Odometry and a Single Downward-Pointing Camera

Stereo imaging is a technique commonly employed for vision-based navigation. For such applications, two images are acquired from different vantage points and then compared using transformations to extract depth information. The technique is commonly used in robotics for obstacle avoidance or for Simultaneous Localization And Mapping, (SLAM). Yet, the process requires a number of image processing steps and therefore tends to be CPU-intensive, which limits the real-time data rate and use in power-limited applications. Evaluated here is a technique where a monocular camera is used for vision-based odometry. In this work, an optical flow technique with feature recognition is performed to generate odometry measurements. The visual odometry sensor measurements are intended to be used as control inputs or measurements in a sensor fusion algorithm using low-cost MEMS based inertial sensors to provide improved localization information. Presented here are visual odometry results which demonstrate the challenges associated with using ground-pointing cameras for visual odometry. The focus is for rover-based robotic applications for localization within GPS-denied environments.

Swank, Aaron J.↗
NASA NTRS1982

Volcanoes and volcanic provinces - Martian western hemisphere

The recognition of some Martian landforms as volcanoes is based on their morphology and geologic setting. Other structures, however, may exhibit classic identifying features to a varying or a less degree; these may be only considered provisionally as having a volcanic origin. Regional geologic mapping of the western hemisphere of Mars from Viking images has revealed many more probable volcanoes and volcanotectonic features than were recognized on Mariner 9 pictures. These abundant volcanoes have been assigned to several distinct provinces on the basis of their areal distribution. Although the Olympus-Tharsis region remains as the principle center of volcanism on Mars, four other important provinces are now also recognized: the lowland plains, Tempe Terra plateau, southern highlands (in the Phaethontis and Thaumasia quadrangles), and a probable ignimbrite province, situated along the highland-lowland boundary in Amazonis Planitia. Volcanoes in any one province vary in morphlogy, size, and age, but volcanoes in each province tend to have common characteristics that distinguish that particular group.

Scott, D. H.↗
NASA NTRS1994

Active vision in satellite scene analysis

In earth observation or planetary exploration it is necessary to have more and, more autonomous systems, able to adapt to unpredictable situations. This imposes the use, in artificial systems, of new concepts in cognition, based on the fact that perception should not be separated from recognition and decision making levels. This means that low level signal processing (perception level) should interact with symbolic and high level processing (decision level). This paper is going to describe the new concept of active vision, implemented in Distributed Artificial Intelligence by Dassault Aviation following a 'structuralist' principle. An application to spatial image interpretation is given, oriented toward flexible robotics.

Naillon, Martine↗
NASA NTRS1973

Task 10: an ERTS Experiment for Mapping Iron Compounds, 1383

The author has identified the following significant results. Preliminary results of a spectral ratioing technique are shown for a region at the southern end of the Wind River Range, Wyoming. Digital ratio graymaps and analog ratio images have been produced for the test site, but ground truth is not yet available for thorough interpretation of these products. ERTS-1 analog ratio images were found generally better than either ERTS-1 single channel images or high altitude aerial photos for the discrimination of vegetation from non-vegetation in the test site region. Some linear geological features smaller than the ERTS-1 spatial resolution are seen as well in ERTS-1 ratio and single channel images as in high altitude aerial photography. Geochemical information appears to be extractable from ERTS-1 data. Good preliminary quantitative agreement between ERTS-1 derived ratios and laboratory derived reflectance ratios of rocks and minerals encourage plans to use lab data as training sets for a simple ratio gating logic approach to automatic recognition maps. Empirical atmospheric corrections indicate that atmospheric corrections are needed to make the ratio method, and possibly other types of data processing, consistent over large geographical and temporal displacements.

Vincent, R. K.↗
NASA NTRS1973

Ratio maps of iron ore deposits Atlantic City district, Wyoming

Preliminary results of a spectral rationing technique are shown for a region at the southern end of the Wind River Range, Wyoming. Digital ratio graymaps and analog ratio images have been produced for the test site, but ground truth is not yet available for thorough interpretation of these products. ERTS analog ratio images were found generally better than either ERTS single-channel images or high altitude aerial photos for the discrimination of vegetation from non-vegetation in the test site region. Some linear geological features smaller than the ERTS spatial resolution are seen as well in ERTS ratio and single-channel images as in high altitude aerial photography. Geochemical information appears to be extractable from ERTS data. Good preliminary quantitative agreement between ERTS-derived ratios and laboratory-derived reflectance ratios of rocks and minerals encourage plans to use lab data as training sets for a simple ratio gating logic approach to automatic recognition maps.

Vincent, R. K.↗
NASA NTRS1975

Diagnostic accuracy of an ultrasonic multiple transducer cardiac imaging system

An ultrasonic multiple-transducer imaging system for intracardiac structure visualization is developed in order to simplify visualization of the human heart in vivo without radiation hazard or invasion of the body. Results of the evaluation of the diagnostic accuracy of the devised system in a clinical setting for adult patients are presented and discussed. Criteria are presented for recognition of mitral valva prolapse, mitral stenosis, pericardial effusion, atrial septal defect, and left ventricular dyssynergy. The probable cause for false-positive and false-negative diagnoses is discussed. However, hypertrophic myopathy and congestive myopathy were unable to be detected. Since only qualitative criteria were used, it was not possible to differentiate patients with left ventricular volume overload from patients without cardiac pathology.

Popp, R. L.↗
NASA NTRS1976

Documentation of procedures for textural/spatial pattern recognition techniques

A C-130 aircraft was flown over the Sam Houston National Forest on March 21, 1973 at 10,000 feet altitude to collect multispectral scanner (MSS) data. Existing textural and spatial automatic processing techniques were used to classify the MSS imagery into specified timber categories. Several classification experiments were performed on this data using features selected from the spectral bands and a textural transform band. The results indicate that (1) spatial post-processing a classified image can cut the classification error to 1/2 or 1/3 of its initial value, (2) spatial post-processing the classified image using combined spectral and textural features produces a resulting image with less error than post-processing a classified image using only spectral features and (3) classification without spatial post processing using the combined spectral textural features tends to produce about the same error rate as a classification without spatial post processing using only spectral features.

Haralick, R. M.↗
NASA NTRS1990

Design of optimal correlation filters for hybrid vision systems

Research is underway at the NASA Johnson Space Center on the development of vision systems that recognize objects and estimate their position by processing their images. This is a crucial task in many space applications such as autonomous landing on Mars sites, satellite inspection and repair, and docking of space shuttle and space station. Currently available algorithms and hardware are too slow to be suitable for these tasks. Electronic digital hardware exhibits superior performance in computing and control; however, they take too much time to carry out important signal processing operations such as Fourier transformation of image data and calculation of correlation between two images. Fortunately, because of the inherent parallelism, optical devices can carry out these operations very fast, although they are not quite suitable for computation and control type operations. Hence, investigations are currently being conducted on the development of hybrid vision systems that utilize both optical techniques and digital processing jointly to carry out the object recognition tasks in real time. Algorithms for the design of optimal filters for use in hybrid vision systems were developed. Specifically, an algorithm was developed for the design of real-valued frequency plane correlation filters. Furthermore, research was also conducted on designing correlation filters optimal in the sense of providing maximum signal-to-nose ratio when noise is present in the detectors in the correlation plane. Algorithms were developed for the design of different types of optimal filters: complex filters, real-value filters, phase-only filters, ternary-valued filters, coupled filters. This report presents some of these algorithms in detail along with their derivations.

Rajan, Periasamy K.↗
NASA NTRS1992

Optical computing and neural networks; Proceedings of the Meeting, National Chiao Tung Univ., Hsinchu, Taiwan, Dec. 16, 17, 1992

The present conference discusses optical neural networks, photorefractive nonlinear optics, optical pattern recognition, digital and analog processors, and holography and its applications. Attention is given to bifurcating optical information processing, neural structures in digital halftoning, an exemplar-based optical neural net classifier for color pattern recognition, volume storage in photorefractive disks, and microlaser-based compact optical neuroprocessors. Also treated are the optical implementation of a feature-enhanced optical interpattern-associative neural network model and its optical implementation, an optical pattern binary dual-rail logic gate module, a theoretical analysis for holographic associative memories, joint transform correlators, image addition and subtraction via the Talbot effect, and optical wavelet-matched filters. (No individual items are abstracted in this volume)

Hsu, Ken-Yuh↗
NASA NTRS2003

Array biosensor for detection of toxins

The array biosensor is capable of detecting multiple targets rapidly and simultaneously on the surface of a single waveguide. Sandwich and competitive fluoroimmunoassays have been developed to detect high and low molecular weight toxins, respectively, in complex samples. Recognition molecules (usually antibodies) were first immobilized in specific locations on the waveguide and the resultant patterned array was used to interrogate up to 12 different samples for the presence of multiple different analytes. Upon binding of a fluorescent analyte or fluorescent immunocomplex, the pattern of fluorescent spots was detected using a CCD camera. Automated image analysis was used to determine a mean fluorescence value for each assay spot and to subtract the local background signal. The location of the spot and its mean fluorescence value were used to determine the toxin identity and concentration. Toxins were measured in clinical fluids, environmental samples and foods, with minimal sample preparation. Results are shown for rapid analyses of staphylococcal enterotoxin B, ricin, cholera toxin, botulinum toxoids, trinitrotoluene, and the mycotoxin fumonisin. Toxins were detected at levels as low as 0.5 ng mL(-1).

Review↗
NASA NTRS1976

Machine processing of remotely acquired data

The two major branches of remote sensing are based on image orientation and numerical orientation. Numerically oriented systems tend to involve computers for data analysis. In designing an information-gathering system, the sensor as well as the means of analysis must be well mated to the type of system orientation. Attention is given to sensor types as related to system types, the multispectral approach and pattern recognition, the multispectral scanner as data source, an illustrative example, procedural details in the use of pattern recognition, the speed and cost of data processing, the use of spatial information, and data preprocessing steps.

Landgrebe, D. A.↗
NASA NTRS1990

Use of laser range finders and range image analysis in automated assembly tasks

A proposition to study the effect of filtering processes on range images and to evaluate the performance of two different laser range mappers is made. Median filtering was utilized to remove noise from the range images. First and second order derivatives are then utilized to locate the similarities and dissimilarities between the processed and the original images. Range depth information is converted into spatial coordinates, and a set of coefficients which describe 3-D objects is generated using the algorithm developed in the second phase of this research. Range images of spheres and cylinders are used for experimental purposes. An algorithm was developed to compare the performance of two different laser range mappers based upon the range depth information of surfaces generated by each of the mappers. Furthermore, an approach based on 2-D analytic geometry is also proposed which serves as a basis for the recognition of regular 3-D geometric objects.

Alvertos, Nicolas↗
NASA NTRS1992

A vision architecture for the extravehicular activity retriever

The Extravehicular Activity Retriever (EVAR) is a robotic device currently being developed by the Automation and Robotics Division at the NASA Johnson Space Center to support activities in the neighborhood of the Space Shuttle or Space Station Freedom. As the name implies, the Retriever's primary function will be to provide the capability to retrieve tools, equipment or other objects which have become detached from the spacecraft, but it will also be able to rescue a crew member who may have become inadvertently de-tethered. Later goals will include cooperative operations between a crew member and the Retriever such as fetching a tool that is required for servicing or maintenance operations. This report documents a preliminary design for a Vision System Planner (VSP) for the EVAR that is capable of achieving visual objectives provided to it by a high level task planner. Typical commands which the task planner might issue to the VSP relate to object recognition, object location determination, and obstacle detection. Upon receiving a command from the task planner, the VSP then plans a sequence of actions to achieve the specified objective using a model-based reasoning approach. This sequence may involve choosing an appropriate sensor, selecting an algorithm to process the data, reorienting the sensor, adjusting the effective resolution of the image using lens zooming capability, and/or requesting the task planner to reposition the EVAR to obtain a different view of the object. An initial version of the Vision System Planner which realizes the above capabilities using simulated images has been implemented and tested. The remaining sections describe the architecture and capabilities of the VSP and its relationship to the high level task planner. In addition, typical plans that are generated to achieve visual goals for various scenarios will be discussed. Specific topics to be addressed will include object search strategies, repositioning of the EVAR to improve the quality of information obtained from the sensors, complementary usage of the sensors and redundant capabilities.

Magee, Michael↗
NASA NTRS1993

A vision system planner for increasing the autonomy of the Extravehicular Activity Helper/Retriever

The Extravehicular Activity Retriever (EVAR) is a robotic device currently being developed by the Automation and Robotics Division at the NASA Johnson Space Center to support activities in the neighborhood of the Space Shuttle or Space Station Freedom. As the name implies, the Retriever's primary function will be to provide the capability to retrieve tools and equipment or other objects which have become detached from the spacecraft, but it will also be able to rescue a crew member who may have become inadvertently de-tethered. Later goals will include cooperative operations between a crew member and the Retriever such as fetching a tool that is required for servicing or maintenance operations. This paper documents a preliminary design for a Vision System Planner (VSP) for the EVAR that is capable of achieving visual objectives provided to it by a high level task planner. Typical commands which the task planner might issue to the VSP relate to object recognition, object location determination, and obstacle detection. Upon receiving a command from the task planner, the VSP then plans a sequence of actions to achieve the specified objective using a model-based reasoning approach. This sequence may involve choosing an appropriate sensor, selecting an algorithm to process the data, reorienting the sensor, adjusting the effective resolution of the image using lens zooming capability, and/or requesting the task planner to reposition the EVAR to obtain a different view of the object. An initial version of the Vision System Planner which realizes the above capabilities using simulated images has been implemented and tested. The remaining sections describe the architecture and capabilities of the VSP and its relationship to the high level task planner. In addition, typical plans that are generated to achieve visual goals for various scenarios are discussed. Specific topics to be addressed will include object search strategies, repositioning of the EVAR to improve the quality of information obtained from the sensors, and complementary usage of the sensors and redundant capabilities.

Magee, Michael↗
NASA NTRS1993

Design of coupled mace filters for optical pattern recognition using practical spatial light modulators

Spatial light modulators (SLMs) are being used in correlation-based optical pattern recognition systems to implement the Fourier domain filters. Currently available SLMs have certain limitations with respect to the realizability of these filters. Therefore, it is necessary to incorporate the SLM constraints in the design of the filters. The design of a SLM-constrained minimum average correlation energy (SLM-MACE) filter using the simulated annealing-based optimization technique was investigated. The SLM-MACE filter was synthesized for three different types of constraints. The performance of the filter was evaluated in terms of its recognition (discrimination) capabilities using computer simulations. The correlation plane characteristics of the SLM-MACE filter were found to be reasonably good. The SLM-MACE filter yielded far better results than the analytical MACE filter implemented on practical SLMs using the constrained magnitude technique. Further, the filter performance was evaluated in the presence of noise in the input test images. This work demonstrated the need to include the SLM constraints in the filter design. Finally, a method is suggested to reduce the computation time required for the synthesis of the SLM-MACE filter.

Rajan, P. K.↗
NASA NTRS1975

A comparative interregional analysis of selected data from LANDSAT-1 and EREP for the inventory and monitoring of natural ecosystems

Comparative statistics were presented on the capability of LANDSAT-1 and three of the Skylab remote sensing systems (S-190A, S-190B, S-192) for the recognition and inventory of analogous natural vegetations and landscape features important in resource allocation and management. Two analogous regions presenting vegetational zonation from salt desert to alpine conditions above the timberline were observed, emphasizing the visual interpretation mode in the investigation. An hierarchical legend system was used as the basic classification of all land surface features. Comparative tests were run on image identifiability with the different sensor systems, and mapping and interpretation tests were made both in monocular and stereo interpretation with all systems except the S-192. Significant advantage was found in the use of stereo from space when image analysis is by visual or visual-machine-aided interactive systems. Some cost factors in mapping from space are identified. The various image types are compared and an operational system is postulated.

Poulton, C. E.↗
NASA NTRS1977

Skylab simulator visual displays and training for joint observing programs

The complexity of the Skylab Apollo Telescope Mount (ATM) experiment operations necessitated the use of high fidelity simulations of the onboard visual displays and pointing system for crew training. The displays which were simulated included the H-alpha displays, XUV monitor display, XUV/slit/white light display, X-ray image display, and the white light coronagraph display. The pointing simulation was achieved by projecting film sequences which were subsequently viewed by TV cameras. An optical system in front of the vidicons simulated the pointing, roll, and zoom capabilities of the ATM and sensing systems. The simulation enabled the Skylab crewmen to obtain valuable integrated training combining such tasks as target recognition and status assessment, complex and the time dependent pointing operations, malfunction analyses, and rapid responses to flare and other transient events.

Holt, A. C.↗