Showing papers by "Azriel Rosenfeld published in 1992"

Computer Vision and Image Processing

Abstract: On the use of morphological operators in a class of edge detectors, L. Hertz and R.W. Schafer a valley-seeking threshold selection technique, S.C. Sahasrabudhe and K.S. Das Gupta local characteristics of binary images and their application to the automatic control of low-level robot vision, P.W. Pachowicz corner detection and localization in a pyramid, S. Baugher and A. Rosenfeld parallel-hierarchical image partitioning and region extraction, G.N. Khan and D.F. Gillies invariant architectures for low-level vision, L. Jacobson and H. Wechsler representation - primitives chain code, L. O'Gorman generalized cones - useful geometric properties, K. Rao and G. Medioni vision-based rendering - image synthesis for vision feature algorithms, J.D. Yates, et al recognition - investigation of a number of character recognition algorithms, A.A. Verikas, et al log-polar mapping applied to pattern representation and recognition, J.C. Wilson and R.M. Hodgson pattern recognition of binary image objects using morphological shape decomposition, I. Pitas and N.D. Sidiropoulos a pattern classification approach to multi-level thresholding for image segmentation, J.G. Postaire and M. Ameziane KOR - a knowledge-based object recognition system, C.M. Lee, et al shape decomposition based on perceptual structure, H.S. Kim and K.H. Park three dimensional - the Frobenius metric in image registration, K. Zikan and T.M. Silberberg binocular fusion revisited utilizing a log-polar tessellation, N.C. Griswold, et al an expert system for recovering 3D shape and orientation from a single view, W.J. Shomar, et al integrating intensity and range sensing to construct 3D polyhedra representation, W.N. Lie, et al notes - texture segmentation using topographic labels, T.C. Pong, et al an improved algorithm for labelling connected components in a binary image, X.D. Yang a note on the paper "The Visual Potential - One Convex Polygon", A. Laurentini a string descriptor for matching partial shapes, H.C. Liu and M.D. Srinath formulation and error analysis for a generalized image point correspondence algorithm, S. Fotedar, et al a new surface tracking system in 3D binary images, L.W. Chang and M.J. Tsai.

3-D shape recovery using distributed aspect matching

Abstract: An approach to the recovery of 3-D volumetric primitives from a single 2-D image is presented. The approach first takes a set of 3-D volumetric modeling primitives and generates a hierarchical aspect representation based on the projected surfaces of the primitives; conditional probabilities capture the ambiguity of mappings between levels of the hierarchy. From a region segmentation of the input image, the authors present a formulation of the recovery problem based on the grouping of the regions into aspects. No domain-independent heuristics are used; only the probabilities inherent in the aspect hierarchy are exploited. Once the aspects are recovered, the aspect hierarchy is used to infer a set of volumetric primitives and their connectivity. As a front end to an object recognition system, the approach provides the indexing power of complex 3-D object-centered primitives while exploiting the convenience of 2-D viewer-centered aspect matching; aspects are used to represent a finite vocabulary of 3-D parts from which objects can be constructed. >

From volumes to views: an approach to 3-D object recognition

Abstract: We present an approach to the recovery and recognition of 3-D objects from a single 2-D image. The approach is motivated by the need for more powerful indexing primitives, and shifts the burden of recognition from the model-based verification of simple image features to the bottom-up recovery of complex volumetric primitives. Given a recognition domain consisting of a database of objects, we first select a set of object-centered 3-D volumetric modeling primitives that can be used to construct the objects. Next, using a CAD system, we generate the set of aspects of the primitives. Unlike typical aspect-based recognition systems that use aspects to model entire objects, we use aspects to model the parts from which the objects are constructed. Consequently, the number of aspects is fixed and independent of the size of the object database. To accommodate the matching of partial aspects due to primitive occlusion, we introduce a hierarchical aspect representation based on the projected surfaces of the primitives; a set of conditional probabilities captures the ambiguity of mappings between the levels of the hierarchy. From a region segmentation of the input image, we present a novel formulation of the primitive recovery problem based on grouping the regions into aspects. No domain dependent heuristics are used; we exploit only the probabilities inherent in the aspect hierarchy. Once the aspects are recovered, we use the aspect hierarchy to infer a set of volumetric primitives and their connectivity relations. Subgraphs of the resulting graph, in which nodes represent 3-D primitives and arcs represent primitive connections, are used as indices to the object database. The verification of object hypotheses consists of a topological verification of the recovered graph, rather than a geometrical verification of image features. A system has been built to demonstrate the approach, and it has been successfully applied to both synthetic and real imagery.

Recovery of temporal information from static images of handwriting

Abstract: A taxonomy of local, regional, and global temporal clues that, along with a detailed examination of the document, allow temporal properties to be recovered from the image is provided. It is shown that this system will benefit from obtaining a comprehensive understanding of the handwriting signal and that it requires a detailed analysis of stroke and sub-stroke properties. It is suggested that this task requires breaking away from traditional thresholding and thinning techniques, and a framework for such analysis is presented. It is shown how the temporal clues can reliably be extracted from this framework and how many of the seemingly ambiguous situations can be resolved by the derived clues and knowledge of the writing process. >

Compact object recognition using energy-function-based optimization

Abstract: Describes a method of recognizing objects whose contours can be represented in smoothly varying polar coordinate form. Both low- and high-level information about the object (contour smoothness and edge sharpness at the low level and contour shape at the high level) are incorporated into a single energy function that defines a 1D, cyclic, Markov random field (1DCMRF). This 1DCMRF is based on a polar coordinate object representation whose center can be initialized at any location within the object. The recognition process is based on energy function minimization, which is implemented by simulated annealing. >

Recognition of planar shapes from perspective images using contour-based invariants

Abstract: This paper presents a new method of recognizing a two-dimensional shape from a single perspective projection image taken from an unknown (three-dimensional) viewpoint. The method is based on quadratic approximations to the effect of the slant of an inverse perspective transformation on angles and lengths. These approximations allow us to define contour-based properties that are invariant under perspective transformation. The method can be used to recognize partially occluded shapes, as well as shapes that are not exactly related by perspective transformations. When a shape is recognized, the method also provides estimates of its tilt and slant.

Perceptual motion transparency : the role of geometrical information

Abstract: Perceptual motion transparency occurs whenever two or more patterns are seen moving at different depth levels, such that we can see one pattern move across the others, and perceptual motion coherence occurs when we see a single motion. We present here a model of perceptual motion transparency and coherence that consists of three stages: (i) measure the normal velocity along contours or the velocity of features such as corners or line end points; (ii) take the intersection, in velocity space, of all possible pairs of constraint lines associated with the normal velocity components; and (iii) combine the results of steps (i) and (ii) in the velocity histogram, which is the plot of the total number of votes for each velocity in velocity space. For two patterns we perceive motion coherence, transparency, or a mixture of both types of motion depending on whether the velocity histogram is unimodal, bimodal, or trimodal. According to our model we perceive motion transparency or coherence depending on the total number of prominent peaks of the velocity histogram, where each peak is located at the position corresponding to the velocity of one of the patterns or of coherent motion. We show that the number of prominent peaks in the velocity histogram depends on the error in the measurement of the local velocity and on the relative orientation of the local velocities along contours; this relative orientation encodes contour shape. Our model differs from current motion theories in that it describes the perception of motion not as a result of local velocity extraction but instead as the result of the integration of local velocity and geometrical shape information across different points of the image and across superimposed patterns. This model allows for the occurrence of mixed motion perception, which arises from the combination of the velocity information associated with motion transparency and coherence. We have tested this model through computational and psychophysical experiments done with line patterns. As a result of these experiments we conjecture that the human visual system may use at least three stages to process image velocity.

Analysis of the least median of squares estimator for computer vision applications

Abstract: The robust least-median-of-squares (LMedS) estimator, which can recover a model representing only half the data points, was recently introduced in computer vision. Image data, however, is usually also corrupted by a zero-mean random process (noise) accounting for the measurement uncertainties. It is shown that in the presence of significant noise, LMedS loses its high breakdown point property. A different, two-stage approach in which the uncertainty due to noise is reduced before applying the simplest LMedS procedure is proposed. The superior performance of the technique is proved by comparative graphs. >

Maximum-likelihood edge detection in digital signals

Abstract: This paper treats the problem of edge detection in noisy piecewise-constant digital signals, using a maximum likelihood approach. Conventional edge detectors usually assume that the noise is Gaussian, and do not take advantage of prior knowledge about the ensemble of signals (aside from the assumption that the signals are piecewise constant). Our approach can handle noise that has an arbitrary probability density function; it also makes use of prior probability densities for the piece sizes and values in the signal ensemble.

Temporal clues in handwriting

Abstract: Handwritten character recognition is typically classified as online or offline depending on the nature of the input data. Online data consists of a temporal sequence of instrument positions while offline data is in the form of a 2D image of the writing sample. Online recognition techniques have been relatively successful but have the disadvantage of requiring the data to be gathered during the writing process. This paper presents work on the extraction of temporal information from static images of handwriting and its implications for character recognition. >

Purposive recognition: an active and qualitative approach

Abstract: We propose an alternative way to study the problem of visual recognition which is closer to the spirit emerging from Brooks' work on building robots than to Marr's reconstructive approach. Our theory is purposive in the sense that recognition is considered in the context of an agent performing it in an environment, along with the agent's intentions that translate into a set of behaviors; it is qualitative in the sense that only partial recovery is needed; it is active in the sense that various partial recovery tasks need for recognition are achieved through active vision; and it is opportunistic in the sense that every available cue is used.© (1992) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Primitive-based shape modeling and recognition

Abstract: We present an approach to the recovery and recognition of 3-D objects from a single 2-D image. Given a recognition domain consisting of a database of objects, we select a set of object-centered 3-D volumetric modeling primitives that can be used to construct the objects. Next, we take the set of primitives and generate a hierarchical aspect representation based on their projected surfaces; conditional probabilities capture the ambiguity of mappings between levels of the hierarchy. From a region segmentation of the input image, we present a novel formulation of the recovery problem based on grouping the regions into aspects. No domain dependent heuristics are used; we exploit only the probabilities inherent in the aspect hierarchy. Once the aspects are recovered, we use the aspect hierarchy to infer a set of volumetric primitives and their connectivity. Subgraphs of the resulting graph, in which nodes represent 3-D primitives and arcs represent primitive connections, are used as indices into the object database. Object verification consists of a topological verification of the recovered graph rather than a geometrical verification of image features.

Object recognition by a robotic agent: the purposive approach

Abstract: Studies the problem of object recognition by considering it in the context of an agent operating in an environment, where the agent's intentions translate into a set of behaviors. In this context, an object can fulfil a function; if the agent recognizes this, it has in effect recognized the object. To perform this type of recognition one needs on one hand a definition of the desired function, and on the other the means of determining whether the object can fulfil that function. To illustrate this approach the authors describe the visual recognition abilities that might be needed by an autonomous cleaning robot. >

Models: the graphics-vision interface

Abstract: Graphics and vision have complementary goals. A central goal of graphics is to generate realistic images of a scene, given a description of the scene. Conversely, the goal of vision is to describe a scene, give a set of real images of the scene. Progress in both fields is closely tied to our ability to model classes of scenes and to use these models efficiently in the synthesis and analysis of images. A “model” is defined, in general, as a constraint on the set of possible scenes; this paper briefly discusses, on a qualitative level, some major classes of models.

Formulation and error analysis for a generalized image point correspondence algorithm

Abstract: A Generalized Image Point Correspondence (GIPC) algorithm, which enables the determination of 3-D motion parameters of an object in a configuration where both the object and the camera are moving, is discussed. A detailed error analysis of this algorithm has been carried out. Furthermore, the algorithm was tested on both simulated and video-acquired data, and its accuracy was determined.

Generation of Random Planar Tessellations by Hierarchical Decimation

Abstract: We describe a method of generating a class of irregular tessellations of a digital image using a random decimation process. The decimation is hierarchical and can generate regions of varying sizes. The tessellation can be generated in O(log image size) time if the process is implemented in parallel on a cellular pyramid computer.