Showing papers on "Orientation (computer vision) published in 1987"

A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-the-shelf TV cameras and lenses

Abstract: A new technique for three-dimensional (3D) camera calibration for machine vision metrology using off-the-shelf TV cameras and lenses is described. The two-stage technique is aimed at efficient computation of camera external position and orientation relative to object reference coordinate system as well as the effective focal length, radial lens distortion, and image scanning parameters. The two-stage technique has advantage in terms of accuracy, speed, and versatility over existing state of the art. A critical review of the state of the art is given in the beginning. A theoretical framework is established, supported by comprehensive proof in five appendixes, and may pave the way for future research on 3D robotics vision. Test results using real data are described. Both accuracy and speed are reported. The experimental results are analyzed and compared with theoretical prediction. Recent effort indicates that with slight modification, the two-stage calibration can be done in real time.

Relations between the statistics of natural images and the response properties of cortical cells.

Abstract: The relative efficiency of any particular image-coding scheme should be defined only in relation to the class of images that the code is likely to encounter. To understand the representation of images by the mammalian visual system, it might therefore be useful to consider the statistics of images from the natural environment (i.e., images with trees, rocks, bushes, etc). In this study, various coding schemes are compared in relation to how they represent the information in such natural images. The coefficients of such codes are represented by arrays of mechanisms that respond to local regions of space, spatial frequency, and orientation (Gabor-like transforms). For many classes of image, such codes will not be an efficient means of representing information. However, the results obtained with six natural images suggest that the orientation and the spatial-frequency tuning of mammalian simple cells are well suited for coding the information in such images if the goal of the code is to convert higher-order redundancy (e.g., correlation between the intensities of neighboring pixels) into first-order redundancy (i.e., the response distribution of the coefficients). Such coding produces a relatively high signal-to-noise ratio and permits information to be transmitted with only a subset of the total number of cells. These results support Barlow's theory that the goal of natural vision is to represent the information in the natural environment with minimal redundancy.

Model-based recognition and localization from sparse range or tactile data

Abstract: This paper discusses how local measurements of three-dimensional positions and surface normals (recorded by a set of tactile sensors, or by three-dimensional range sensors), may be used to identify and locate objects from among a set of known objects. The objects are modeled as polyhedra having up to six degrees of freedom relative to the sensors. We show that inconsistent hypotheses about pairings between sensed points and object surfaces can be discarded efficiently by using local constraints on distances between faces, angles between face normals, and angles (relative to the surface normals) of vectors between sensed points. We show by simulation and by mathematical bounds that the number of hypotheses consistent with these constraints is small We also show how to recover the position and orientation of the object from the sensory data. The algorithm's performance on data obtained from a triangulation range sensor is illustrated.

Error modeling in stereo navigation

Abstract: In stereo navigation, a mobile robot estimates its position by tracking landmarks with on-board cameras. Previous systems for stereo navigation have suffered from poor accuracy, in part because they relied on scalar models of measurement error in triangulation. Using three-dimensional (3D) Gaussian distributions to model triangulation error is shown to lead to much better performance. How to compute the error model from image correspondences, estimate robot motion between frames, and update the global positions of the robot and the landmarks over time are discussed. Simulations show that, compared to scalar error models, the 3D Gaussian reduces the variance in robot position estimates and better distinguishes rotational from translational motion. A short indoor run with real images supported these conclusions and computed the final robot position to within two percent of distance and one degree of orientation. These results illustrate the importance of error modeling in stereo vision for this and other applications.

The cortex transform: rapid computation of simulated neural images

Abstract: With a goal of providing means for accelerating the image processing, machine vision, and testing of human vision models, an image transform was designed, which makes it possible to map an image into a set of images that vary in resolution and orientation. Each pixel in the output may be regarded as the simulated response of a neuron in human visual cortex. The transform is amenable to a number of shortcuts that greatly reduce the amount of computation.

Orthogonal Pyramid Transforms For Image Coding.

Abstract: We describe a set of pyramid transforms that decompose an image into a set of basis functions that are (a) spatial-frequency tuned, (b) orientation tuned, (c) spatially localized, and (d) self-similar. For computational reasons the set is also (e) orthogonal and lends itself to (f) rapid computation. The systems are derived from concepts in matrix algebra, but are closely connected to decompositions based on quadrature mirror filters. Our computations take place hierarchically, leading to a pyramid representation in which all of the basis functions have the same basic shape, and appear at many scales. By placing the high-pass and low-pass kernels on staggered grids, we can derive odd-tap QMF kernels that are quite compact. We have developed pyramids using separable, quincunx, and hexagonal kernels. Image data compression with the pyramids gives excellent results, both in terms of MSE and visual appearance. A non-orthogonal variant allows good performance with 3-tap basis kernels and the appropriate inverse sampling kernels.

Relative Orientation

Abstract: Before corresponding points in images taken with two cameras can be used to recover distances to objects in a scene, one has to determine the position and orientation of one camera relative to the other. This is the classic photogrammetric problem of {\it relative orientation}, central to the interpretation of binocular stereo information. Described here is a particularly simple iterative scheme for recovering relative orientation that, unlike existing methods, does not require a good initial guess for the baseline and the rotation.

Physically Based Simulation Model for Acoustic Sensor Robot Navigation

Abstract: A computer model is described that combines concepts from the fields of acoustics, linear system theory, and digital signal processing to simulate an acoustic sensor navigation system using time-of-flight ranging. By separating the transmitter/receiver into separate components and assuming mirror-like reflectors, closed-form solutions for the reflections from corners, edges, and walls are determined as a function of transducer size, location, and orientation. A floor plan consisting of corners, walls, and edges is efficiently encoded to indicate which of these elements contribute to a particular pulse-echo response. Sonar maps produced by transducers having different resonant frequencies and transmitted pulse waveforms can then be simulated efficiently. Examples of simulated sonar maps of two floor plans illustrate the performance of the model. Actual sonar maps are presented to verify the simulation results.

Techniques for calibration of the scale factor and image center for high accuracy 3D machine vision metrology

Abstract: This paper describes techniques for calibrating certain intrinsic camera parameters for machine vision. The parameters to be calibrated are the horizontal scale factor, i.e. the factor that relates the sensor element spacing of a discrete array camera to the picture element spacing after sampling by the image acquisition circuitry, and the image center, i.e. the intersection of the optical axis with the camera sensor. The scale factor calibration uses a 1D-FFT and is accurate and efficient. It also permits the use of only one coplanar set of calibration points for general camera calibration. Three groups of techniques for center calibration are presented: Group I requires using a laser and a four-degree of freedom adjustment of its orientation, but is simplest in concept, and is accurate and reproducible. Group II is simple to perform, but is less accurate than the other two. The most general Group III is accurate and efficient, but requires accurate image feature extraction of calibration points with known 3D coordinates. A feasible setup is described. Results of real experiments are presented and compared with theoretical predictions. Accuracy and reproducibility of the calibrated parameters are reported, as well as the improvement in actual 3D measurement due to center calibration.

Generating an interpretation tree from a CAD model for 3D-object recognition in bin-picking tasks

Abstract: This article describes a method to generate 3D-object recognition algorithms from a geometrical model for bin-picking tasks. Given a 3D solid model of an object, we first generate apparent shapes of an object under various viewer directions. Those apparent shapes are then classified into groups (representative attitudes) based on dominant visible faces and other features. Based on the grouping, recognition algorithms are generated in the form of an interpretation tree. The interpretation tree consists of two parts: the first part for classifying a target region in an image into one of the shape groups, and the second part for determining the precise attitude of the object within that group. We have developed a set of rules to find out what appropriate features are to be used in what order to generate an efficient and reliable interpretation tree. Features used in the interpretation tree include inertia of a region, relationship to the neighboring regions, position and orientation of edges, and extended Gaussian images. This method has been applied in a task for bin-picking objects that include both planar and cylindrical surfaces. As sensory data, we have used surface orientations from photometric stereo, depth from binocular stereo using oriented-region matching, and edges from an intensity image.

Special purpose automatic programming for 3D model-based vision

Abstract: A method for the automatic construction of fast special purpose vision programs is described. The starting point for the automatic construction process is a description of a particular 3D object. The result is a fast special purpose program for recognizing and locating that object in images, without restriction on the orientation of the object in space. The method has been implemented and tested on a variety of images with good results. Some of the tests involved images in which the target objects appear in a jumbled pile. The current implementation is not fully optimized for speed. However, evidence is given that image analysis times on the order of a second or less can be obtained for typical industrial recognition tasks. (This time estimate excludes edge finding).

New Methods for Matching 3-D Objects with Single Perspective Views

Abstract: In this paper we analyze the ability of a computer vision system to derive properties of the three-dimensional (3-D) physical world from viewing two-dimensional (2-D) images. We present a new approach which consists of a model-based interpretation of a single perspective image. Image linear features and linear feature sets are backprojected onto the 3-D space and geometric models are then used for selecting possible solutions. The paper treats two situations: 1) interpretation of scenes resulting from a simple geometric structure (orthogonality) in which case we seek to determine the orientation of this structure relatively to the viewer (three rotations) and 2) recognition of moderately complex objects whose shapes (geometrical and topological properties) are provided in advance. The recognition technique is limited to objects containing, among others, straight edges and planar faces. In the first case the computation can be carried out by a parallel algorithm which selects the solution that has received the largest number of votes (accumulation space). In the second case an object is uniquely assigned to a set of image features through a search strategy. As a by-product, the spatial position and orientation (six degrees of freedom) of each recognized object is determined as well. The method is valid over a wide range of perspective images and it does not require perfect low-level image segmentation. It has been successfully implemented for recognizing a class of industrial parts.

Measuring Visual Motion From Image Sequences

Abstract: Motion is an important and fundamental source of visual information. It is well known that the pattern of image motion contains information useful for the determination of the 3-dimensional structure of the environment and the relative motion between the camera and the objects in the scene. However, the accurate measurement of image motion from a sequence of real images has proven to be difficult. In this thesis, a hierarchical framework for the computation of dense displacement fields from pairs of images, and an integrated system consistent with that framework are described. Each input intensity image is first decomposed using a set of spatial-frequency tuned channels. The information in the low-frequency channels is used to provide rough displacements over a large range, which are then successively refined by using the information in the higher-frequency channels. Within each channel, a direction-dependent confidence measure is computed for each displacement vector, and a smoothness constraint is used to propagate reliable displacement vectors to their neighboring areas with less reliable vectors. For our integrated system, Burt's Laplacian pyramid transform is used for the spatial-frequency decomposition, and the minimization of the sum of squared differences measure (SSD) is used as the match criterion. The confidence measure is derived from the shape of the SSD surface, and the smoothness constraint is formulated as a functional minimization problem. Results of applying our system to several image-pairs containing complex camera motion as well as independently moving objects are included. A number of well-known gradient-based and matching techniques are also shown to be consistent with our framework. The mathematical relationship between the gradient-based techniques and a class of correlation techniques is established. This thesis also includes several proposals for extending our approach for multiple-frame analysis. Of particular interest is an approach which involves the decomposition of the input images according to orientation as well as scale. This new approach unifies the spatio-temporal energy models, which are currently popular in psychophysics, with the gradient-based and the matching techniques, and appears biologically feasible, and ideally suited for connectionist models of computation.

Residual vision in a scotoma. A follow-up study of 'form' discrimination.

Abstract: This study is a follow-up of a patient, D.B., who was reported (Weiskrantz et al., 1974) to be able to discriminate between simple visual forms within the scotoma caused by a lesion in calcarine cortex. Among other capacities, he was able to discriminate between lines of different orientation in the frontal plane. Given the reported deficits for form discrimination but a high sensitivity for orientation discrimination in primates without striate cortex, the question arises whether D.B.'s apparent 'form' discrimination arises from an ability to discriminate the orientations of components of the figures. It is shown in the first experiment, by using the optic disc as a control, that his ability to detect stimuli in his scotoma cannot be due to stray light falling upon the intact field. Next, his ability to discriminate orientations is confirmed, and an orientation threshold determined. A range of form discriminations is presented varying in degree of orientation cues of their components. It is confirmed that he can discriminate those forms originally studied, in which such differences are large, but not when orientation cues are small or minimal. Finally, his ability to compare two forms both projected within his scotoma is examined. Even when the components of the two forms have large orientation differences and are highly discriminable when presented successfully, D.B. appears to be unable to make a 'same-different' comparison when both are presented simultaneously. The evidence is interpreted against D.B.'s having a residual capacity for form discrimination.

Shape recovery and segmentation with deformable part models

Abstract: A method for recovery of compact volumetric models for shape representation and segmentation in computer vision is introduced. The models are superquadrics with parametric deformations (bending, tapering, and cavity deformation). The input for the model recovery is three-dimensional range points. We define an energy or cost function whose value depends on the distance of points from the model's surface and on the overall size of the model. Model recovery is formulated as a least-squares minimization of the cost function for all range points belonging to a single part. The initial estimate required for minimization is the rough position, orientation and size of the object. During the iterative gradient descent minimization process, all model parameters are adjusted simultaneously, recovering position, orientation, size and shape of the model, such that most of the given range points lie close to the model's surface. Because of the ambiguity of superquadric models, the same shape can be described with different sets of parameters. A specific solution among several acceptable solutions, which are all minima in the parameter space, can be reached by contraining the search to a part of the parameter space. The many shallow local minima in the parameter space are avoided as a solution by using a stochastic technique during minimization. Segmentation is defined as a description of objects or scenes in terms of the adopted shape vocabulary. Model recovery of an object consisting of several parts starts by computing the rough position, orientation and size of the whole object. By allowing a variable number of range points in a model, a model can actively search for a better fit (by compressing itself and expanding) resulting in a subdivision of the object into a model representing the largest part of the object and points belonging to the rest of the scene. Using the same method, the remaining points can be recursively subdivided into parts each represented with a single compact volumetric model. Results using real range data show that the recovered models are stable and that the recovery procedure is fast.

Qualitatne landmark-based path planning and following

Abstract: This paper develops a theory for path planning and following using visual landmark recognition for the representation of environmental locations. It encodes local perceptual knowledge in structures called viewframes and orientation regions. Rigorous representations of places as visual events are developed in a uniform framework that smoothly integrates a qualitative version of path planning with inference over traditional metric representations. Paths in the world are represented as sequences of sets of landmarks, viewframes. orientation boundary crossings, and other distinctive visual events. Approximate headings are computed between view frames that have lines of sight to common landmarks. Orientation regions are range-free, topological descriptions of place that are rigorously abstracted from viewframes. They yield a coordinate-free model of visual landmark memory that can also be used for path planning and following. With this approach, a robot can opportunistically observe and execute visually cued "shortcuts".

Manipulation of run-length encoded images

Abstract: A method and apparatus for manipulating run-length encoded rasterized images. Sizing, slanting, rotating or otherwise transforming an image outline to a new orientation is accomplished without converting image information into a bit map or discrete pixel format. An image outline is characterized in terms of visible and invisible vectors along an input raster scan line by relating run-lengths in a previous input scan line with run-lengths in a current input scan line. The resulting vector characterization allows determination of crossover points on output raster scan lines for the manipulated image by means of transform coefficients. Memory bins store these crossover points, and these bins are sorted to construct a new run-length encoded image outline.

Electronic-automatic orientation device for walkers and the blind

Abstract: The automatic orientation device according to the invention for walkers and the blind is a small light weight and easily man-carried instrument. It contains an electronic-digital-calculating for continually determining the vectorial sum of the steps made by converting these steps into electrical pulses proportional to the strides and by automatic determination of the direction by means of a direction emitter in order to calculate the distance and direction covered from the point of departure. Whether by walking, running or horse riding the user is thus provided with optical or acoustic indicating or with perceptible vibrations indicating the distance and direction from the point of departure. In this way any person, including the blind, can find his way to the point of departure without any means of assistance, or geographical map or town plan, and in particular when the visibility is extremely poor.

Finding Texture Boundaries in Images

Abstract: : Texture provides one cue for identifying the physical cause of an intensity edge, such as occlusion, shadow, surface orientation or reflectance change. Marr, Julesz, and others have proposed that texture is represented by small lines or blobs, called 'textons' by Julesz (1981a), together with their attributes, such as orientation, elongation, and intensity. Psychophysical studies suggest that texture boundaries are perceived where distributions of attributes over neighborhoods of textons differ significantly. However, these studies, which deal with synthetic images, neglect to consider two important questions: How can these textons be extracted from images of natural scenes? And how, exactly, are texture boundaries then found? This thesis proposes answers to these questions by presenting an algorithm for computing blobs from natural images and a statistic for measuring the difference between two sample distributions of blob attributes. As part of the blob detection algorithm, methods for estimating image noise are presented, which are applicable to edge detection as well.

Delineating Elliptical Objects with an Application to Cardiac Scintigrams

Abstract: To delineate the myocardium in planar thallium-201 scintigrams of the left ventricle, a method, based on the Hough transformation, is presented. The method maps feature points (X, Y, Y')-where Y' reflects the direction of the tangent in edge point (X,Y)-into the two-dimensional space of the axis lengths of the ellipse. Within this space, a probability density function (pdf) can be estimated. When the center of the ellipse or its orientation are unknown, the 2-D pdf of the lengths of the axes is extended to a 5-D pdf of all parameters of the ellipse (lengths of the axes, coordinates of the center, and the orientation). It is shown that the variance of the edge-point-based estimates of the axis lengths increases when the location error of the center of the supposed ellipse or its orientation error increases. The likelihood of the estimates is expected to decrease with increasing variance. Therefore, local search algorithms can be applied to find the maximum likelihood estimate of the parameters of the ellipse. Curves describing the convergency of the algorithm are presented, as well as an example of the application of the algorithm to real scintigrams. The method is able to detect contours even if they are only partly visualized, as in thallium scintigrams of the myocardium of patients with ischemic heart disease. As long as the number of parameters describing the contour is relatively low, such an algorithm is also suitable for application to differently curved contours.

Automatic assembly apparatus

Abstract: In automatic assembly apparatus operating under electronic vision control, features on components to be assembled must be recognized and located in position and orientation so that a gripper mechanism may be directed to one component to pick it up and move it into correct position and orientation relative to the other component for assembly. The invention provides a known parallel projection optical system 8, 9, at the location of each component which provides a plan view for an electronic camera 11 of the components 20, 22 at constant scale regardless of lateral or axial component movements. The grey-level picture of each component provided by the camera is thresholded into a binarized picture at a threshold level which selects a primary component feature 27 within a part of the camera field of view which is certain to contain this feature. From the known location of secondary features 25, 26 and 38, 39, 40 of the component relative to the primary feature, successively limited search areas 35 and 41, 42 are set up within the camera field of view to select these secondary features when thresholded at levels related to the primary threshold. Sufficient features are thereby located in the camera field of view to provide position and orientation information on the component to a computer which directs the gripper 15 to assemble the components.

TINA: the sheffield AIVRU vision system

Abstract: We describe the Sheffield AIVRU 3D vision system for robotics. The system currently supports model based object recognition and location; its potential for robotics applications is demonstrated by its guidance of a UMI robot arm in a pick and place task. The system comprises: 1) The recovery of a sparse depth map using edge based passive stereo triangulation. 2) The grouping, description and segmentation of edge segments to recover a 3D description of the scene geometry in terms of straight lines and circular arcs. 3) The statistical combination of 3D descriptions for the purpose of object model creation from multiple stereo views, and the propagation of constraints for within view refinement. 4) The matching of 3D wireframe models to 3D scene descriptions, to recover an initial estimate of their position and orientation.

Adaptive Enhancement Based On A Visual Model

Abstract: The human visual system is capable of detecting and following the course of striated periodic patterns, even under adverse conditions of poor contrast and low signal-to-noise ratio. Sections of a striated pattern of subthreshold contrast may be detected easily if other parts of the same pattern have suprathreshold contrast. To simulate these capabilities of the visual system, an image processing algorithm was developed using basic "cells" that are well localized in both the space and spatial frequency domains. These band-limiting, orientation-sensitive "fan filters" are similar in their point spread functions to the two-dimensional Gabor functions commonly used to describe responses of visual cortical cells. These filters are used both to detect the orientation of the striated pattern in a small window and to enhance the image in that orientation. The search for local orientation is limited to a small range based on orientations found in neighboring, overlapping windows. The orientation of the maximally responding cell is used for the enhancement. Results of applying the adaptive directional enhancement to nerve fiber layer photographs, finger-prints, and seismic data are presented.

Transformation of image positions, rotations, and sizes into shift parameters.

Abstract: An optical image processing system is described that converts orientation and size to shift properties and simultaneously preserves the positional information as a shift. The system is described analytically and experimentally. The transformed image can be processed further with a classical correlator working with a rotational and size-invariant. multiplexed match filter. An optical robot vision system designed on this concept would be able to look at several objects simultaneously and determine their shape, size, orientation, and position with two measurements on the input scene at different rotation angles.

Line detection in digital pictures: a hypothesis prediction/verification pardigm

Abstract: We present an efficient method for detecting straight line segments in digital pictures using a hypothesis prediction/verification paradigm. In this paradigm, a straight line segment of predefined length is predicted to exist at some particular pixel location. The orientation of this predicted line segment is based on the edge orientation at the pixel location. This prediction is then verified against statistical tests performed on the line. As a result, the predicted line is either validated as being a line segment, or it is rejected. An extension of this algorithm for the detection of lines at different lengths is also presented, and a criterion is defined in order to evaluate the significance of the detected line segments.

Sensitivity to corners in flow patterns

Abstract: Flow patterns are two-dimensional orientation structures that arise from the projection of certain kinds of surface coverings (such as fur) onto images. Detecting orientation changes within them is an important task, since the changes often correspond to significant events such as corners, occluding edges, or surface creases. We model such patterns as random-dot Moire patterns, and examine sensitivity to change in orientation within them. We show that the amount of structure available from which orientation and curvature can be estimated is critical, and introduce a path-length parameter to model it. For short path lengths many discontinuities are smoothed over, which has further implications for computational modeling of orientation selectivity.

Minimization of directed points generated in three-dimensional dividing cubes method

Abstract: A system for displaying three-dimensional surface structures according to computer graphics methods extracts a surface definition from a tomographic array of data using interpolation of the data for smooth, high resolution images. Interpolation can be performed to a degree where artifact-free images are produced for all viewing orientations. Data-processing capacity and time requirements can be reduced with less interpolation while image quality is maintained for all viewing orientations by inspecting the viewing orientation and appropriately scaling the image.

TINA: A 3D Vision System for Pick and Place.

Abstract: The paper describes the Sheffield AIVRU 3D vision system for robotics. The system currently supports model-based object recognition and location; its potential for robotics applications is demonstrated by its guidance of a UMI robot arm in a pick-and-place task. The system comprises: recovery of a sparse depth map using edgebased passive stereo triangulation; grouping, description and segmentation of edge segments to recover a 3D description of the scene geometry in terms of straight lines and circular arcs; statistical combination of 3D descriptions for the purpose of object model creation from multiple stereo views, and the propagation of constraints for within-view refinement; and matching 3D wireframe models to 3D scene descriptions to recover an initial estimate of their position and orientation.

Model-based knowledge-based optical processors

Abstract: An efficient 3-D object-centered knowledge base is described. The ability to on-line generate a 2-D image projection or range image for any object/viewer orientation from this knowledge base is addressed. Applications of this knowledge base in associative processors and symbolic correlators are then discussed. Initial test results are presented for a multiple degree of freedom object recognition problem. These include new techniques to achieve object orientation information and two new associative memory matrix formulations.