A system based on passive stereo machine vision is conceptually explored as a new approach to the problem of obstacle warning/avoidance. Past systems considered for obstacle warning have included microwave (radar), ac...
详细信息
A system based on passive stereo machine vision is conceptually explored as a new approach to the problem of obstacle warning/avoidance. Past systems considered for obstacle warning have included microwave (radar), acoustical, and active (lidar) visible/infrared methods. However passive stereo machine vision has distinct advantages over all of these, primarily because it provides more information for scene understanding algorithms than other types of systems. A new software package named ATOM™ (for Automatic TOpographic Mapper) would be capable of extracting ranges at distances of 500 ft. and greater with sufficient accuracy to determine closing velocities of obstacles. If the computation cycle time were as low as 0.5 sec, this would be adequate to warn a driver in sufficient time to come to a complete halt before hitting stationary obstacles on dry or wet roads. Although computation cycle time on an off-the-shelf parallel processing computer is estimated at approximately 22 seconds, a computation time of 0.5 seconds is expected to be possible in two years on new versions of the same computer.
Data fusion of multispectral image data requires techniques that are often time-consuming while giving unclear results. Development of algorithms that integrate information in a useful way is important to improving au...
详细信息
Data fusion of multispectral image data requires techniques that are often time-consuming while giving unclear results. Development of algorithms that integrate information in a useful way is important to improving autonomous and semi-autonomous image understanding systems. This paper presents a comparison of two data fusion methods, each of which compresses the data. One method, the Hotelling transform (Karhunen-Loeve transform), is investigated and its results compared with a less computationally intensive method using new techniques. Each algorithm is translated into the Air Force's Image Algebra, as it provides a common mathematical environment for image algorithm development, optimization, comparison, coding and performance evaluation. The algebra facilitates the comparison of the advantages and disadvantages of each method.
This paper describes an approach to collision avoidance for rotorcraft flying at low altitudes. The method is based on extracting information from passive image-based sensors. These methods can be used to alert a pilo...
详细信息
This paper describes an approach to collision avoidance for rotorcraft flying at low altitudes. The method is based on extracting information from passive image-based sensors. These methods can be used to alert a pilot about obstacles and provide inputs to navigation and guidance systems for further processing.
In the past decade, vision/sensing systems techniques have received significant attention for applications in robotics and automation. The thrust of automation and robotics for space applications has been proposed for...
详细信息
In the past decade, vision/sensing systems techniques have received significant attention for applications in robotics and automation. The thrust of automation and robotics for space applications has been proposed for increased productivity, improved reliability, increased flexibility, higher safety, and for the performance of tasks unsuited to humans. These benefits can be insured through automating time-consuming tasks, increasing productivity/performance of crew-accomplished tasks, and performing tasks beyond the capability of the crew. This paper provides a review of efforts currently in progress at the NASA/Johnson Space Center and at Rice University, the accomplishments to date, and some of the anticipated future developments. Both systems and algorithms are discussed. The evolution of future vision/sensing is projected to include the fusion of multisensors ranging from microwave to optical with multimode capability to include position, attitude, recognition, and motion parameters. The algorithms for information extraction would incorporate aspects of intelligence and knowledge for the interpolation and extrapolation of the needed data. The key features of the overall system design will be small size and weight, fast signal processing, robust algorithms, and accurate parameter determination. These aspects of vision sensing will also be discussed in this paper.
Image processing techniques (and papers describing them) have evolved rapidly in recent years. Lately, publications in image processing have underscored advances in compression, restoration, and enhancement of gray le...
详细信息
Image processing techniques (and papers describing them) have evolved rapidly in recent years. Lately, publications in image processing have underscored advances in compression, restoration, and enhancement of gray level and color coded image information using dedicated pipeline processing and systolic array architectures. These new algorithms and specialized hardware have been developed and applied to such real-time tasks as automated testors on manufacturing lines and robotic vehicle guidance systems. Often overlooked is the part of binary image processing. A review of binary processing techniques is valuable in understanding the the basic operating principles of high-speed optical inspection devices in particular and many pipeline array processing devices in general. The class of binary image transforms covered here was originally developed for topological operations on threshold image data. These non-linear operations are based on boolean transformations of a pixel and its eight adjacent neighbors (in a discrete Cartesian plane) and are named morphological transforms by authors. A straightforward way to visualize such transforms is outlined in the paper and will be demonstrated at the conference. The demonstration program package was written using the IBM Personal Computer BASI language and was presented as an educational tool in a graduate level course in image processing. The skeletal program included in the text was enhanced to allow real-time demonstration of these transform techniques and encourages readers to experiment and develop their own lookup tables and image transforms. Understanding the function and generation of these lookup tables, which can provide any arbitrary boolean combination of the nine binary values available from a pixel and its eight Cartesian neighbors, is the key to grasping the image-transforms illustrated here. Examples in the paper and presentation give the reader an idea of the power and flexibility these transforms provide in im
Exact characterization of the array in terms of its geometry and the sensor gain and phase is important in beamforming and directions-of-arrival estimation problems. In practice the sensor gain and phase may not be kn...
详细信息
opticalinformationprocessing research aimed at Space Station automation applications is reviewed. The emphasis of the NASA Ames Research Center program is on intelligent optical pattern recognition and optical contr...
详细信息
opticalinformationprocessing research aimed at Space Station automation applications is reviewed. The emphasis of the NASA Ames Research Center program is on intelligent optical pattern recognition and optical control processing. Attention is given to the primary functions of an overall scene understanding system: distortion-invariant optical feature generation, dimensionality reduction, object classification, and contextual informationprocessing. A method of using synthetic discriminant functions to facilitate learning in a high-speed optical correlator is described. A discussion is presented of candidate analog and digital architectures for the optical implementation of state-estimation algorithms needed for the control of high-dimension dynamic systems. The multivariate system chosen for the optical control technology demonstration--a segmented, adaptive mirror and interferometrically based wavefront sensor--is also described.
A novel approach to the general problem of signal parameter estimation is described. Though the technique (ESPRIT) is discussed in the context of direction-of-arrival estimation, it can be applied to a wide variety of...
详细信息
The estimation of wavefront aberrations is central to both optical testing and real time control of active opticalsystems. optical measurements, whether wavefront tilts, optical path differences, or image plane inten...
The estimation of wavefront aberrations is central to both optical testing and real time control of active opticalsystems. optical measurements, whether wavefront tilts, optical path differences, or image plane intensities, must be numerically transformed into meaningful aberrations. This dissertation develops new, more accurate and more noise tolerant algorithms to process these measurements. The text begins by presenting a number of generic estimation techniques suitable for processing measurements taken at a single point in time. All of these static processing techniques prove to be variations on the Bayesian minimum variance estimator. Reviews of a point spread function measurement processor and a Hartman sensor processor serve as an introduction to practical aberration measurement processing. Two approaches to testing afocal telescopes with an array of small flats instead of the usual single large flat, called subaperture testing, are analyzed in more detail. Then a new subaperture testing processor is developed which utilizes information on the correlations among subaperture misalignments. Subsequent coding and testing with computer simulated measurements show that the new processor is significantly more accurate than classical algorithms. The static aberration measurement theory is then extended to accommodate sequences of measurements taken over a number of points in time. A new subaperture testing measurement processor using a Kalman filter is developed in order to take advantage of the dynamic nature of these measurements. Coding and testing with computer simulated measurements for a number of test cases with known aberrations, show that the new processor is considerably more accurate than any previous method.
暂无评论