A connectionist approach for peak detection in Hough space
TL;DR: A connectionist network is presented for detecting peaks in multidimensional Hough space and the neural network implementation successfully uses circumstantial evidence and detects multiple winners over the parameter space such that these winners correspond to parameters of features in the image.
About: This article is published in Pattern Recognition.The article was published on 1992-10-01. It has received 8 citations till now. The article focuses on the topics: Hough transform & Feature detection (computer vision).
Citations
More filters
TL;DR: It is shown that for multiband images, multithresholding subsets of bands followed by a fusion stage results in improved performance and running time.
155 citations
Cites methods from "A connectionist approach for peak d..."
...[27] was actually proposed for peak picking in the Hough accumulator via an arti®cial neural networkbased algorithm....
[...]
TL;DR: An accurate and robust evaluator that dynamically removes contributions of backgrounds and analyzes voting patterns around peaks in the accumulator space is proposed that can be applied to any HT that keeps the basic characteristics of the voting process.
111 citations
Cites methods from "A connectionist approach for peak d..."
...Neural network was designed for peak detection in multi-dimensional parameter space [16]....
[...]
TL;DR: In this article, a processus d'extraction des toits de bâtiments a partir d'images aeriennes ou d'image satellite a haute-resolution is presented, which se base sur les caracteristiques propres a ces zones urbaines regularisees.
Abstract: Le developpement rapide de certaines villes se fait parfois avec des techniques de construction industrialisees. Cette solution conduit a la creation de vastes zones urbanisees, dites «regularisees», ou l'on ne retrouve qu'un petit nombre de modeles de bâtiments. On presente dans cet article un processus d'extraction des toits de bâtiments a partir d'images aeriennes ou d'images satellite a haute-resolution qui se base sur les caracteristiques propres a ces zones urbaines regularisees. Ce processus adopte une demarche ascendante et l'on insiste sur le fait qu'il reduit le nombre des details de bas-niveau qui sont pris en compte lors de l'extraction du contour des toitures. On utilise a cet effet un ensemble de regles geometriques, definies au travers de modeles de bâtiments. Des essais effectues sur des scenes reelles ont montre que cette methode fournissait une detection efficace des coins a angle droit, ameliorant globalement la qualite de l'extraction.
32 citations
Cites background from "A connectionist approach for peak d..."
...It should be noted that this threshold determination approach is based on evidence for the existence of a line in image space, not on the relative neighbourhood of the suspected peak in the accumulator space (for a general purpose peak detection approach in Hough space, see Vinod et al. (1992))....
[...]
TL;DR: A two-layer neural-network model is designed which accepts image coordinates as the input and learns the parametric form of conoidal shapes (lines/circles/ellipses) adaptively, which not only reduces the large space requirements of the classical Hough transform, but also represents parameters with a higher precision.
Abstract: A two-layer neural-network model is designed which accepts image coordinates as the input and learns the parametric form of conoidal shapes (lines/circles/ellipses) adaptively. It provides an efficient representation of visual information embedded in the connection weights and the parameters of the processing elements. It not only reduces the large space requirements of the classical Hough transform (HT), but also represents parameters with a higher precision. The performance of the methodology is compared with other existing algorithms and has been found to excel over those algorithms in many cases.
18 citations
Cites background from "A connectionist approach for peak d..."
...In order to do so, the quantized parameter space needs to be maintained in some form depending on the type of the network (in most of the cases including the self-organizing models [28], [31], it is in the form of an array of neurons)....
[...]
TL;DR: In this paper, the authors describe a general approach for detecting data alignments in large unordered noisy multidimensional datasets, which is independent of the geometric properties of the alignments to be detected.
17 citations
References
More filters
TL;DR: A model of a system having a large number of simple equivalent components, based on aspects of neurobiology but readily adapted to integrated circuits, produces a content-addressable memory which correctly yields an entire memory from any subpart of sufficient size.
Abstract: Computational properties of use of biological organisms or to the construction of computers can emerge as collective properties of systems having a large number of simple equivalent components (or neurons). The physical meaning of content-addressable memory is described by an appropriate phase space flow of the state of a system. A model of such a system is given, based on aspects of neurobiology but readily adapted to integrated circuits. The collective properties of this model produce a content-addressable memory which correctly yields an entire memory from any subpart of sufficient size. The algorithm for the time evolution of the state of the system is based on asynchronous parallel processing. Additional emergent collective properties include some capacity for generalization, familiarity recognition, categorization, error correction, and time sequence retention. The collective properties are only weakly sensitive to details of the modeling or the failure of individual devices.
16,652 citations
TL;DR: It is shown how the boundaries of an arbitrary non-analytic shape can be used to construct a mapping between image space and Hough transform space, which makes the generalized Houghtransform a kind of universal transform which can beused to find arbitrarily complex shapes.
4,310 citations
TL;DR: A new conceptual framework and a minimization principle together provide an understanding of computation in model neural circuits that represent an approximation to biological neurons in which a simplified set of important computational properties is retained.
Abstract: A new conceptual framework and a minimization principle together provide an understanding of computation in model neural circuits. The circuits consist of nonlinear graded-response model neurons organized into networks with effectively symmetric synaptic connections. The neurons represent an approximation to biological neurons in which a simplified set of important computational properties is retained. Complex circuits solving problems similar to those essential in biology can be analyzed and understood without the need to follow the circuit dynamics in detail. Implementation of the model with electronic devices will provide a class of electronic circuits of novel form and function.
2,019 citations
TL;DR: An historical discussion is provided of the intellectual trends that caused nineteenth century interdisciplinary studies of physics and psychobiology by leading scientists such as Helmholtz, Maxwell, and Mach to splinter into separate twentieth-century scientific movements.
1,586 citations
TL;DR: In this article, it was shown that limit sets of such systems cannot be more complicated than invariant sets of systems of one lower dimension, and orthogonal projection along any positive direction maps a limit set homeomorphically and equivariantly onto an invariant set of a Lipschitz vector field in a hyperplane.
Abstract: A vector field in n-space determines a competitive (or cooperative) system of differential equations provided all the off-diagonal terms of its Jacobian matrix are nonpositive (or nonnegative). The principal result is that limit sets of such systems cannot be more complicated than invariant sets of systems of one lower dimension. In fact orthogonal projection along any positive direction maps a limit set homeomorphically and equivariantly onto an invariant set of a Lipschitz vector field in a hyperplane. Limit sets are nowhere dense, unknotted and unlinked. In dimension 2 every trajectory is eventually monotone. In dimension 3 a compact limit set which does not contain an equilibrium is a closed orbit or a cylinder of closed orbits. Introduction. One of the most interesting questions to ask about a dynamical system is: what is the long-run behavior of its trajectories? In many systems it is natural to expect, or at least hope, that almost all trajectories either converge to an equilibrium or asymptotically approach a closed orbit (= periodic trajectory). Unfortu- nately there are many systems that not only lack this convenient property, but cannot even be approximated by systems that have it. Such systems are often said to be "chaotic" or to possess "strange attractors". To make matters worse, it is very hard to discover the long-run behavior of any but the simplest systems. Research on this problem has bifurcated into two quite different methodologies. A great deal of recent work has gone toward exploring the consequences of various assumptions about the large scale structure of the system, e.g., hyperbolicity of the nonwandering set, structural stability, ergodicity, and so forth. The basic examples come from geometry and physics; the mathematical tech- niques tend to be topological. For a recent overview of this work see Smale (15, Chapt. I).
389 citations