Showing papers on "Orientation (computer vision) published in 1980"

Photometric Method For Determining Surface Orientation From Multiple Images

Abstract: A novel technique called photometric stereo is introduced. The idea of photometric stereo is to vary the direction of incident illumination between successive images, while holding the viewing direction constant. It is shown that this provides sufficient information to determine surface orientation at each image point. Since the imaging geometry is not changed, the correspondence between image points is known a priori. The technique is photometric because it uses the radiance values recorded at a single image location, in successive views, rather than the relative positions of displaced features. Photometric stereo is used in computer-based image understanding. It can be applied in two ways. First, it is a general technique for deter-mining surface orientation at each image point. Second, it is a technique for determining object points that have a particular surface orientation. These applications are illustrated using synthesized examples.

Interpreting line drawings as three-dimensional surfaces

Abstract: We propose a computational model for interpreting line drawings as three-dimensional surfaces, based on constraints on local surface orientation along extremal and discontinuity-boundaries Specific techniques are described for two key processes: recovering the three-dimensional conformation of a space curve (eg, a surface boundary) from its two-dimensional projection in an image, and interpolating smooth surfaces from orientation constraints along extremal boundaries

Visual Information Processing: The Structure and Creation of Visual Representations

Abstract: For human vision to be explained by a computational theory, the first question is plain: What are the problems that the brain solves when we see? It is argued that vision is the construction of efficient symbolic descriptions from images of the world. An important aspect of vision is therefore the choice of representations for the different kinds of information in a visual scene. An overall framework is suggested for extracting shape information from images, in which the analysis proceeds through three representations: (1) the primal sketch, which makes explicit the intensity changes and local two-dimensional geometry of an image; (2) the 2 1/2-D sketch, which is a viewer-centred representation of the depth, orientation and discontinuities of the visible surfaces; and (3) the 3-D model representation, which allows an object-centred description of the three-dimensional structure and organization of a viewed shape. The critical act in formulating computational theories for processes capable of constructing these representations is the discovery of valid constraints on the way the world behaves, that provide sufficient additional information to allow recovery of the desired characteristic. Finally, once a computational theory for a process has been formulated, algorithms for implementing it may be designed, and their performance compared with that of the human visual processor.

The influence of vision on the absolute identification of sound-source position

Abstract: The influence of vision on auditory localization was assessed in an absolute identification paradigm using sighted and blindfolded subjects. Vision improved the accuracy of judgments directly in front of, to the side of, and behind the head of subjects in the horizontal plane, but had little relevance to vertical-plane localization. The exact form of the observed facilitation depended on the orientation of the speaker array to the head. In a second experiment involving sound localization in 10 visual environments, there was evidence for the operation of two distinct influences of vision on directional hearing. One result supported the hypothesis that vision provides a frame of reference for judgments, and a second indicated the importance of vision to the maintenance of spatial memory.

Modelling the Environment of an Exploring Vehicle by Means of Stereo Vision

Abstract: This dissertation describes research involving vision techniques which would be useful in an autonomous exploring vehicle, such as a Mars rover. These techniques produce a description of the surroundings of the vehicle in terms of the position, size, and approximate shape of objects, and can match such scene descriptions with others previously produced. The information produced is thus useful both for navigation and obstacle avoidance. The techniques operate by using three-dimensional data which they can produce by means of stereo vision from stereo picture pairs or which can be obtained from a laser rangefinder. The research thus divides conveniently into two portions: stereo mapping and three-dimensional modelling and matching. The stereo mapping techniques are designed to be suitable for the kind of pictures that a Mars rover might obtain and to produce the kind of data that the modelling techniques need. These stereo techniques are based upon area correlation and produce a depth map of the scene. Emphasis is placed upon extraction of useful data from noisy pictures and upon the estimation of the accuracy of the data produced. Included are the following: a self-calibration method for computing the stereo camera model (the relative position and orientation of the two camera positions); a high-resolution stereo correlator for producing accurate matches with accuracy and confidence estimates, which includes the ability to compensate for brightness and contrast changes between the pictures; a search technique for using the correlator to produce a dense sampling of matched points for a pair of pictures; and the computation of the distances to the matched points, including the propagation of the accuracy estimates. The three-dimensional modelling and matching techniques are designed to be tolerant of the errors that stereo mapping techniques often produce. First, a ground surface finder tries to find a set of points that form a well-defined smooth surface that lies below most of the other points. Then, by using this knowledge of the ground surface and knowledge of the camera viewpoint that produced the points in the scene, an object finder approximates the objects that are above the ground by ellipsoids. Finally, a scene matcher can use the descriptions of scenes in terms of ellipsoidal objects. By using a search pruned by using probabilities obtained by means of Bayes' theorem, it determines the probability that two scene descriptions refer to the same scene and the linear transformation needed to bring the two scenes into alignment. These techniques have been tried on stereo pictures of the Martian surface taken by the Viking Lander 1. The object finder was able to locate rocks fairly successfully, and the scene matcher was able to match successfully the resulting scene descriptions. Examples of these results are shown.

Surface Perception from Local Analysis of Texture and Contour

Abstract: : The visual analysis of surface shape from texture and surface contour is treated within a computational framework. The aim of this study is to determine valid constraints that are sufficient to allow surface orientation and distance (up to a multiplicative constant) to be computed from the image of surface texture and of surface contours. The report is in three parts. Part I consists of a review of major theories of surface perception, a discussion of vision as computation and of the nature in which three-dimensional information is manifest in the image, and a study of the representation of local surface orientation. Part II reconisiders the familiar 'texture gradient'. Part III examines our preception of surface contours. It is shown that surface contours are strong sources of information about local surface shape.

Hybrid optical and electronic image processing for strain measurements by speckle photography

Abstract: A Young's fringe processor for speckle photographs is described. It automatically measures fringe separation and orientation with an accuracy of 1% and 1 degrees , respectively. The fringe evaluation is fast and computer controlled. Fringe densities of one to twenty fringes within the diffraction pattern can be treated with the same resolution. Application to displacement and strain measurements by focused image speckle photography is reported. Local strains can be determined with an accuracy of ~E = 10(-5) for displacements between 10 and 100 microm and a spatial resolution of 10 mm.

Extracting oblique planes from serial CT sections.

Abstract: Although geometric principles describing planes oblique to an orthogonal image data set are well understood, no implementation has been offered for their practical specification, extraction, and display in a clinical environment. Fast image generation and ease of user specification-requisite credentials for successful clinical implementations-are handicapped by the large volume of data to process. Other difficulties further complicate an interactive solution. Once oblique planes are generated, their orientation is often difficult to perceive without visual cues that aid their registration with standard image formats. In addition, Moire patterns introduced by digital aliasing often currupt resultant views. In this paper, techniques are outlined for simplifying oblique plane specification, a methodology is presented for image construction, and an interactive approach is illustrated to register images for such general view planes. Finally, digital aliasing of oblique planes is discussed, and a solution is given for this application.

Mapping image properties into shape constraints: skewed symmetry, affine-transformable patterns, and the shape-from-texture paradigm

Abstract: In this paper we demonstrate two new approaches to deriving three-dimensional surface orientation information (“shape”) from two-dimensional image cues. The two approaches are the method of affine-transformable patterns and the shape-from-texture paradigm. They are introduced by a specific application common to both: the concept of skewed symmetry. Skewed symmetry is shown to constrain the relationship of observed distortions in a known object regularity to a small subset of possible underlying surface orientations. Besides this constraint, valuable in its own right, the two methods are shown to generate other surface constraints as well. Some applications are presented of skewed symmetry to line drawing analysis, to the use of gravity in shape understanding, and to global shape recovery.

Experience with the Generalized Hough Transform

Abstract: : The Hough Transform is a method for detecting curves by exploiting the duality between points on a curve and parameters of that curve The initial work showed how to detect both analytic curves (Hough, 1962; Duda and Hart, 1972) and non-analytic curves (Merlin and Farber, 1975), in the case of binary edge images This work was generalized to the detection of some analytic curves in grey level images, specifically lines (O'Gorman and Clowes, 1973), circles (Kimme et al, 1975), and parabolas (Wechsler and Sklansky, 1977) Recently, the Hough technique has been extended to the detection of arbitrary non-analytic shapes in grey level images (Ballard), 1979) This shape detection scheme has been implemented and tested on a variety of artificial images and has found application in the analysis of real aerial images Experience to date indicates that the technique is robust with respect to occlusions, but requires reliable edge-element orientation determination (Author)

Recognition and fuzzy description of sides and symmetries of figures by computer

Abstract: The detection of sides and corners is useful and important in the shape perception of visual picture patterns. Similarly, the recognition of symmetry in pictures is helpful in reducing the storage space, in retrieving a missing portion, or in correcting the orientation of its shape. In the present paper, algorithms are found to detect the aides and symmetry of simply closed two-dimensional man-made and machine-generated outlines. A description based on the fuzzy set theoretic approach is also developed for imperfect non-geometric figures and results of the execution of the algorithms are presented.

Using Digital Terrain Data To Model Image Formation In Remote Sensing

Abstract: Computer-based image analysis requires explicit models of the image-forming process in order to deal with the effects of variations in viewing direction, incident illumination, surface slope and surface material. A fixed illumination, surface material and imaging geometry is incorporated into a single model, called a reflectance map, that allows observed brightness to be written as a function of surface orientation. The reflectance map is used to generate synthetic images from digital terrain models. Synthetic images are used to predict properties of real images. This technique is illustrated using Landsat imagery. Accurate shadow regions are determined from a digital terrain model by calculating which surface elements are visible from the light source. Once shadows are determined, the effect of sky illumination and atmospheric haze is estimated.© (1980) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Method and an arrangement for depicting a half-tone image

Abstract: A method and an arrangement for depicting half-tone images of equal-sized image dots provides an arrangement of the image dots in a grid having three directions of orientation at an angle of 60°. Each image dot is thus equidistant from the six directly adjacent image dots. A constant number of adjacent grid positions is always used to form an image patch. The number of image dots in an image patch determines the grey step in question. The filling of the image patches with image dots according to the grey step to be produced always takes place following the same rule. The image dots associated with a particular grey step are equidistant from corresponding image dots in the six directly adjacent image patches.

Application of moments to the analysis of panoramic astronomical photographs

Abstract: Panoramic astronomical photographs may contain information on up to one million images ofstars and galaxies. To extract this information in a day requires a high -speed measuringmachine and image analysis system. Each stellar or galaxian image may be regarded as a con-nected set of pixels above some local threshold. The moments (unit -weighted and intensity -weighted) up to second order of the pixel distribution for each image are calculated andform the basis of a set of image parameters. This type of analysis is suitable for thegreat majority of images which do not show complex structure intrinsic to the source. Thesemoments provide data in three basic categories of information relevant to the astronomer,i.e., astrometric, photometric and image classification. Application of this scheme and theaccuracy of the image parameters is demonstrated with data obtained from the COSMOS measur-ing machine.IntroductionThe information content of a photograph taken by a modern Schmidt telescope is enormous.Given that the resolution of a fine grain IIIaJ or IIIaF emulsion is 200 lines per mm, thena photograph of size 356 x 356 mm2 contains 5 x 109 resolution elements. These photographsare capable of reaching magnitude E = 23 and may record up to 1 million images of stars andgalaxies.To extract information on this number of images requires a high -speed measuring machineand image analysis system. The COSMOS (Co- Ordinates, Size, Magnitude, Orientation andShape) machine at Edinburgh has been described previously.? Image detection is achieved bysmoothing the variable background and applying a threshold cut at a fixed percentage levelin intensity units above the local background. The background following, background smooth-ing and thresholding procedures have been described by Martin and Lutz.2 Each stellar orgalaxian image is considered as a set of connected pixels above the local threshold level.The pattern analysis algorithm for connecting these pixels to form a coherent image isdescribed by Lutz,3,4This paper will discuss the type of image analysis carried out and give examples of theapplication to astronomical photographs and the resultant accuracy obtained.Moment analysisIn analysing a direct photograph the basic information in which an astronomer is inter-ested occurs in three categories - astrometric, photometric and image classification.Analysis by calculating the moments of the image pixel distribution provides information inall three categories. It is well known that the moments of a function completely anduniquely describe that function. In practice only moments up to second order are calcu-

The orientation of components for automatic assembly

Abstract: Complex components are often difficult to feed and orientate. In this article the use of vision for detecting component orientation is presented, based on a photodiode array and a microprocessor system.

Scale and orientation in combined Doppler and triangulation nets

Abstract: In a combined Doppler and terrestrial net adjustment not only the known systematic discrepancies in scale and orientation between the Doppler measurements and the terrestrial results must be modelled, but also all available informations about the accuracy of these systematic differences are to be taken into account. Using the Helmert-block method for the combination procedure, no covariance matrices for the terrestrially determined coordinates must be computed, their numerical evaluation being a computational detour. The proposed procedure as applied to real nets, includes all different kinds of geometric or physical models, whereby their specific parameters are eliminated at this level. Two solutions are discussed, a three-dimensional and a two-dimensional one, but “two-dimensional” is not equivalent to “non-spatial” in this context.

Automatic orientation control of industrial manipulator - has video camera monitoring orientation of component for comparison with required position

Abstract: An orientation control system for use with industrial manipulators is designed to allow components to be gripped in an arbitrary position followed by automatic positioning in the required location. A component (2) is held in the jaws (8) of an industrial manipulator (7) and is scanned by a video camera (1). The image produced is of varying contrast and is interpreted as an actual position location image (4). The image is compared (6) with reference data (5) stored on magnetic tape. The generated error signal is used to adjust the position of the manipulation.

A statistical technique for recovering surface orientation from texture in natural imagery

Abstract: A statistical method is reported for inferring the shape and orientation of irregularly marked surfaces using image geometry. The basis for solving this problem lies in an understanding of projective geometry, coupled with simple statistical models of the contour generating process. This approach is first applied to the special case of surfaces known to be planar. The distortion of contour shape imposed by projection is treated as a signal to be estimated, and variations of non-projective origin are treated as noise. The resulting method is next extended to the estimation of curved surfaces, and applied successfully to natural images. The statistical estimation strategy is then experimentally compared to human perception of orientation: human observers' judgements of tilt correspond closely to the estimates produced by the planar strategy.

Separation Discrimination in a Simulated Air Traffic Control Display

Abstract: Terminal area controllers made separation judgments concerning whether displayed pairs of aircraft were more than or less than the standard separation of 3 nautical miles (nm) apart The pairs of aircraft varied with respect to their location on the air map, their relative orientation, and their separation Two different map scales (40- and 55-nm radius) were used, and the controllers made 630 separation judgments on each map The psychometric functions obtained indicated that the mean increment in separation required for discrimination was 014 nm There were no significant differences in the psychometric functions for the 40- and 55-nm maps Small, but operationally insignificant, effects on estimates of the separation between displayed aircraft were related to the location of the aircraft on the air map and to the relative orientation of targets

Biological applications of a computer programme package (DIRECT) for statistical analysis of two-dimensional directional data

Abstract: DIRECT, a computer programme package for the analysis of two-dimensional directional data, is descriptionbed. Several examples are given to demonstrate how DIRECT can be used to test hypotheses concerning preferential orientation or direction of movement by animals to external stimuli such as wind, light, rainfall, etc

A Stereophotogrammetric Method For The Measurement Of Body Surfaces Using A Projected Grid

Abstract: A simplified stereophotogrammetric method for the measurement of body surfaces is presented. The advantage of this method in comparison with common stereophotogrammetry is that only one photograph has be evaluated. The technical equipment consists of a projector with a grid slide and a camera. This is equivalent to the common arrangement for stereophotography, if the directions of the light rays emerging from the projector are thought of having been reversed. So the grid slide and the photograph form some sort of a stereo image pair. The grid is projected on the surface to be measured. The evaluation of the photographs is accomplished by the identification of corresponding points. The correspondance can easily be established by counting the crosspoints of the grid. The coordinates of the grid points in the slide can be calculated by this count, whereas those in the photograph have to be measured. Thus a correlation of two images by stereoscopic vision is not necessary. The measurement can be performed with a simple di-gitizer.The calibration of the apparatus is done by using landmarks and solving the normal equations for the orientation elements of the camera and of the projector. With a slight modification the method can be employed for quasi-continuous measurements of the body surface along projected grid lines.© (1980) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

3-D Model Matching For Missile Guidance

Abstract: A range sensor computes the distance from the sensor to the nearest scene point along a given ray. A range image is an array of range values for a raster of ray displacements. Range images preserve the 3-D geometry of a scene as viewed from the sensor. Thus actual measurements of scene geometry, e.g. lengths, areas, etc. can be derived from a range image and compared to similar measurements taken from a prepared reference model. The Lockheed Signal Processing Laboratory has developed a method for utilizing range imagery in intelligence, guidance, and recognition tasks, with particular emphasis on missile guidance using onboard reference imagery. In our approach, the reference model is used to predict which planar surfaces will be visible from the estimated sensor position. Descriptions of these surfaces are stored as a reference list. The sensed range image is then acquired and the local 3-D orientation of each pixel is computed. Adjacent pixels appearing to lie in the same surface are aggregated into primitive planes. Descriptions of these planes form a sensed plane list. The sensed plane list is matched to the reference list in 3-D coordinate space. This matching determines the actual sensor position resulting in an accurate vehicle position fix.© (1980) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Image enhancement by spectral rationing

Abstract: : In a reflectivity image, contrast produced by differences in surface material is often of more interest than that produced by differences in surface orientation or illumination The latter factors may be difficult to predict or to allow for in practice We show that, in certain cases, contrast caused by differences in surface orientation and illumination may be reduced or eliminated by dividing an image obtained in one spectral band by one obtained in a different spectral band We illustrate this technique by enhancing the images of military vehicles recorded on color film, and by enhancing man-made objects in an aerial multispectral scanner image relative to vegetation (Author)

A Digital Video Tracking System

Abstract: The application of image processing and pattern recognition techniques to object identification and tracking at video rates is a problem of wide interest. Previous attempts have been limited to simple thresholding or correlations within a restricted window. New high-speed algorithms have been implemented using micropro-grammable (bit-slice) processors and specialized digital logic to produce an "intelligent" imaging tracker. Adaptive statistical clustering and projection based classification algorithms are applied in real time to identify and track objects that change in appearance through complex and nonstationary background/foreground situations. The system comprises four pipelined processors which, in turn, separate the target image from the background, locate and describe the target shape, establish an intelligent tracking strategy, and generate control signals which drive the telescope mount and optimize the size and orientation of the output image and the gain of the image tube.

3-D Viewer for interpretation of multiple scan sections

Abstract: A new viewing device is being constructed which will allow a physician to examine multiple scan sections simultaneously in their proper orientation in all three dimensions. Test images already produced on a laboratory prototype viewer clearly show paralax and stereoscopic depth cues characteristic of a real three dimensional object. Our experience viewing X-ray CT and ultrasound sections has shown that certain anatomical and pathological features important in cancer detection are more readily appreciated on the 3-D viewer than on conventional planar imaging facilities such as at T.V. monitor or film. The viewer can also form 3-D graphical figures with lines oriented from left-to-right, top-to-bottom and front-to-back. By moving one's head, distant lines appear to move relative to nearby lines, just as would be expected.

A High-Performance Integrated True Graphic Processor

Abstract: The first soon available true graphic display processor (GDP) will be presented It displays an up to 512 × 512 portion of a 4 096 × 4 096 pixel image Vector drawing speed reaches 560 ns/pixel and 96 variable size and orientation characters are available

The Application of Two Dimensional Moment Invariants to Image Signal Processing and Pattern Recognition.

Abstract: : This thesis investigates the application of two-dimensional moment invariants to image pattern recognition. The general problem studied is how to identify an aircraft target and its orientation in real time. The method of moment invariants provides a clever feature extraction technique to reduce the information in an image to a finite number of quantities which are translation, size, and rotation independent.

Image multiplexing with image retrieval using a slit source

Abstract: An image modulated by a random spatial carrier is multiply recorded with a suitable displacement between successive exposures. If a slit source is used for illumination, the visibility of the fringes produced by such a recording depends upon the orientation of the slit with respect to the inter-exposure displacement. This fact has been made use of in the proposed method where the images to be multiplexed have their corresponding inter-exposure displacements lying in different directions. The images are retrieved using suitable binary filters whose transmission characteristics determine the cross-talk between the reconstructed images. A theoretical analysis of the method is presented.