D.H. Ballard, "Generalizing the Hough Transform to Detect Arbitrary Shapes," Pattern Recognition, 13, pp. 111--122 (1981). Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
First 50 documents Next 50
A Highly Adaptable Concept For Visual Inspection - Sablatnig (1997) (Correct)
....algorithms can be used as well. Circle detection To detect circles and circular arcs in the intensity image, we use the Hough transformation [KIE92,YUE90] The Hough method [HOU62] was extended to circle detection by Duda and Hart [DUD75] and extended by Ballard and Brown using the gradients [BAL81]. The detection process has the following steps: 1. Edge detection: The derivatives in the x and y direction are computed and form the gradient; different gradient operators may be used; 2. Hough transform: Circle centers are clusters in the accumulator space; 3. Peak enhancement: The ....
D.H. Ballard, C.M. Brown, "Generalizing the Hough Transform to Detect Arbitrary Shapes", Pattern Recognition, Vol.13, No.2, pp.111-122, 1981.
Automatic Image Segmentation and Classification Using On-line.. - Lee, Street (2000) (1 citation) (Correct)
....invariance is achieved through the use of an object centered coordinate system. Shapes are normalized for size, orientation and re ection during the algorithm, as described in Section 3.3. This representation, motivated by its ease of use template matching using a generalized Hough transform (GHT) [1], allows the matching of a wide variety of topologically simple shapes. 3. Description of the system Figure 2 shows a owchart of the proposed system. The system consists of four parts: detecting, segmenting, clustering and classifying. While these four parts are performed repeatedly, the system ....
....created. The initial templates are created automatically but could easily be drawn by the user. In the second application, we assume no knowledge of the shapes, so the initial templates are created by the user outlining the rst few objects with a mouse. 3.2. Detecting and Segmenting Objects GHT [1] is a typical algorithm for detecting objects in images. Generally, an array of points, of the same size as the input image, comprises the 2D accumulator, in which each point has a value specifying the possibility that the reference point of an object shape to be detected is located at the point. ....
D. H. Ballard. Generalizing the Hough transform to detect arbitrary shapes. Pattern Recognition, 13(2):111-122, 1981.
A Qualitative Profile-based Approach to Edge Detection - Yen (Correct)
....(p, q) in the parameter space, the number of its occurrences is the number of points that are on the line y = px q. Therefore, the higher the number, the more likely this straight line is an edge. Other objects used are circles and ellipses. Arbitrary shapes can be used in Hough transform as well[5], though this generalized Hough transform is used for matching predefined shapes rather than edge detection. In general, Hough transform is used in linking straight lines and circles only. The Hough transform has some advantages such as: 1. It can handle occlusion e#ectively. 2. It is relatively ....
D.H. Ballard. Generalizing the hough transform to detect arbitrary shapes. Pattern Recognition, 13(2):111--122, 1981.
Robust Matching by Dynamic Space Warping for Accurate Face.. - Sahbi, Boujemaa (2001) (Correct)
....present in the scenes. This is due to influences such as the pose, lighting and occlusion. The main idea here is to extract invariant features with respect to these effects, and to compute a similarity between the descriptor derived from the model and the target image features. Many authors ([2], 3]and [4] tackled the problem using simple and compact features like edges which are detectable under a wide variety of pose and lighting conditions. But the main problem in this representation is the difficulty in determining which feature in an image corresponds to each feature on an ....
D. H. Ballard, "Generalizing the hough transform to detect arbitrary shapes," Pattern Recognition, vol. 13, no. 2, pp. 111--122, 1981.
Video-Assisted Global Positioning in Terrain Navigation.. - Le, Seetharaman (Correct)
....the discretization could become a source of inaccuracies in the final outcome. The toroids are generated by incrementing the angles t and , whereas the search space is divided along the X, Y, and Z axes. The error analysis of such generalized of Hough transform techniques could be found in [17]. It is of practical concern to address the possibility that the XYZ volume to be covered by the accumulators may be too large. This could require more computational effort to solve the original problem than the previously known methods to compute the camera position. However, this potential ....
D.H. Ballard. Generalizing the Hough Transform to Detect Arbitrary Shapes. Pattern Recognition, Vol. 13 (No. 2):111-122, 1981.
Fuzzy Hough Transform and an MLP with Fuzzy Input/Output for.. - Sural (1999) (Correct)
....necessary to incorporate fuzzy feature extraction concepts in a neural network. Our approach combines the robustness of feature extraction with the speed of operation of neural networks in a framework of fuzzy systems. Hough transform is a method for the extraction of lines and curves from images [1,22]. Fuzzy probabilistic concepts have been introduced by some authors to generalize the basic Hough transform technique [3] A fuzzy Hough transform method has been presented in [6] where an image point is treated as a fuzzy point. We propose a Hough transform technique in which a number of fuzzy ....
D.H.Ballard, Generalizing the Hough transform to detect arbitrary shapes, Pattern Recognition 13 (1981) 111-122.
Probabilistic Search for Object Segmentation and Recognition - Hillenbrand, Hirzinger (2002) (Correct)
....exploit here are between four data points, that is, tetrahedron geometries, where three of the points are selected from a spherical neighborhood of the fourth. For four points x 1 , x 2 , x 3 , c we define the map #(c; x 1 , x 2 , x 3 ) # r R 4a 3 # 3 4a 3 # 3(r # [0, 1] [0, 1] 13) where r is the mean distance of the center c to x 1 , x 2 , x 3 , R is the radius of the spherical neighborhood of c, d is the (signed) distance of c to the plane defined by x 1 , x 2 , x 3 , and a is the area of the triangle they define . Explicitly, r = 1 x c ....
....are between four data points, that is, tetrahedron geometries, where three of the points are selected from a spherical neighborhood of the fourth. For four points x 1 , x 2 , x 3 , c we define the map #(c; x 1 , x 2 , x 3 ) # r R 4a 3 # 3 4a 3 # 3(r # [0, 1] [0, 1] , 13) where r is the mean distance of the center c to x 1 , x 2 , x 3 , R is the radius of the spherical neighborhood of c, d is the (signed) distance of c to the plane defined by x 1 , x 2 , x 3 , and a is the area of the triangle they define . Explicitly, r = 1 x c , 14) d = ....
[Article contains additional citation context not shown here]
Ballard, D. H. Generalizing the Hough transform to detect arbitrary shapes. Pattern Recognit. 13 (1981), 111--122.
Adaptive Time-Varying Cancellation of Wideband.. - Barbarossa, Scaglione (1999) (3 citations) (Correct)
....onto the time frequency plane using some TFD (e.g. Wigner Ville distribution, etc. and 2) pattern recognition applied to TFD to recognize the signal signature expressed through its instantaneous frequency curve. The tool used for pattern recognition is the generalized Hough transform (HT) [4] for its capability of detecting arbitrary shapes. The method was named Wigner Hough transformation (WHT) 5] Given a signal , its WHT is defined as WHT (5) where denotes the Wigner Ville distribution of (6) and is the interference instantaneous frequency (7) For example, when dealing with ....
D. Ballard, "Generalizing the Hough transform to detect arbitrary shapes," Pattern Recognit., vol. 13, no. 2, pp. 111--122, 1981.
An Efficient Randomized Algorithm for Detecting Circles - Chen, Chung (2001) (1 citation) (Correct)
....methods for detecting circles have been developed. One type of method decomposes the parameter space into many parameter spaces with lower dimension [22] Another type of method uses the gradient information of each edge pixel to reduce the computing time or the requirement of the accumulator [1, 4, 14, 22]. A third type of method uses the geometry property in the circle to improve the performance [9, 10] However, these three types of methods still need some amount of computing time and at least a 2 D accumulator array. Some other recent HT based variants for detecting circles can be found in [12, ....
D. H. Ballard, Generalizing the HoughTransform to detect arbitrary shapes, Pattern Recog. 13, 1981, 111--122.
Recupero di immagini tramite trasformata di Hough - Anelli, Micarelli, Sangineto (2002) (Correct)
....dagli archivi di dati visivi: sarebbe impensabile associare manualmente ad ogni immagine in rete delle didascalie testuali. Nell articolo proponiamo un sistema di retrieval per somiglianza di forma (query by shape) basato su una versione innovativa della Trasformata di Hough Generalizzata (GHT) [1] adatta agli scopi del VIR. La GHT classica, dato un modello , rappresentato tramite una tavola di riferimento (R Table) e un immagine , traduce la ricerca di in in un problema di individuazione dei massimi locali in un accumulatore 25540 aggiornato mediante un meccanismo di ....
D H Ballard, 'Generalizing the Hough Transform to detect arbitrary shapes', Pattern Recognition, 13. No. 2,111-122(1981).
An Analysis of the Effect of Gaussian Error in Object Recognition - Sarachik (1994) (Correct)
....features. The size of the search space is polynomial, O(n ) to be precise. This overall method has come to be associated with Huttenlocher and Ullman ( HU87] who dubbed it alignment , though other previous work used transformation space search (for example, the Hough transform method [Bal81] as well as [Bai84, FB80, TM87] and others) One of the contributions of Huttenlocher s work was to show that a feature pairing of size 3 was necessary and sufficient to solve uniquely for the model pose, and how to do it. Another characteristic of the alignment method as presented in [Hut88] was ....
D.H. Ballard. Generalizing the Hough Transform to Detect Arbitrary Shapes. Pattern Recognition, 13(2), 1981.
An Analysis of the Effect of Gaussian Error in Object Recognition - Sarachik (Correct)
....features. The size of the search space is polynomial, O(n ) to be precise. This overall method has come to be associated with Huttenlocher and Ullman ( HU87] who dubbed it alignment , though other previous work used transformation space search (for example, the Hough transform method [Bal81] as well as [Bai84, FB80, TM87] and others) One of the contributions of Huttenlocher s work was to show that a feature pairing of size 3 was necessary and sufficienttosolve uniquely for the model pose, and how to do it. Another characteristic of the alignment method as presented in [Hut88] was ....
D.H. Ballard. Generalizing the Hough Transform to Detect Arbitrary Shapes. Pattern Recognition, 13(2), 1981.
3D Pose from 3 Corresponding Points under Weak-Perspective.. - Alter (1992) (Correct)
.... correspondences (e.g. 5] 10] 1] 9] 28] 29] 15] 3] 16] 18] 30] 19] In addition, pose clustering techniques use every correspondence between a minimal set of model and image features to compute a model pose, and then count the number of times each pose is repeated (e.g. [2], 26] 25] 23] 11] 4] For computing poses of 3D objects from 2D images, a model of projection must be selected, and typically either perspectiveor weak perspective projection is chosen. Weak perspective projection is an orthographic projection plus a scaling, whichserves to approximate ....
Ballard, D. H., "Generalizing the Hough Transform to Detect Arbitrary Shapes," Pattern Recognition,vol. 13, no. 2, pp. 111-122, 1981.
Illumination Invariance and Object Model in Content-Based .. - Li, Zaïane, Tauber (1999) (1 citation) (Correct)
....of S (edge separation) and # (directionality) The texture statistics are extracted for each locale; in other words, they are locale based. They are derived from the edge image of the luminance image Y, where Y 0.299R 0.587G 0.114B. 3. Shape. The generalized Hough transform (GHT) [31] is adopted to represent the shape of the object. Briefly, each edge point in the object model is represented by a vector r i (# i , r i ) connecting the edge point to a chosen reference point for the object. All r i s are stored in an R table which serves as an object model. The R table is ....
....histogram intersection, i.e. by taking the sum of the minimum of the texture histograms, min H i , H i ( k, # ) 6) If # threshold # 0 , then the color hypothesis has the texture support. The implementation of the generalized Hough transform (GHT) is fairly straightforward [31], except 1. the GHT is only performed on a portion of the database image at the location containing all of the matched locales to save time; 2. the voting for the GHT is only for the single scale and rotation k and # . After the GHT voting, the accumulate array is smoothed (i.e. every 5 5 ....
D. Ballard, Generalizing the Hough transform to detect arbitrary shapes, Pattern Recog. 13(2), 1981, 111--122.
Characterization of Self-Similar Multifractals with Wavelet.. - Hwang, Mallat (1994) (4 citations) (Correct)
....the singular support. We verify this result for fractional Brownian motions. Most multifractals encountered in physics or image processing are not exactly self similar. The deviations from selfsimilarity can be interpreted as a renormalization noise. Voting procedures such as the Hough transform [5], have been particularly successful to recover parameterized curves from noisy data. We introduce such a voting scheme to estimate the renormalization parameters of the wavelet transform maxima curves. Section 4 gives numerical results for a Cantor measure, a dyadique multifractal function, and a ....
....renormalization maps. In most cases, multifractals are approximatively self similar, which means that we must take into account renormalization errors. Many algorithms that estimate the parameters of noisy curves in an image plane are voting procedures in the corresponding parameter space [5]. We describe such a voting algorithm that can recovers non exact renormalization properties in multifractals by using the information provided by all the wavelet transform maxima. Ane renormalizations are characterized by the three parameters (p; l; r) of equation (7) We introduce a voting ....
D. Ballard, \Generalizing the Hough transform to detect arbitrary shapes", Pattern Recognition, vol. 13, no. 2, p. 111-122, 1981.
Deriving Stopping Rules for the Probabilistic Hough.. - Shaked, Yaron, Kiryati (1996) (7 citations) (Correct)
....name for a broad family of algorithms. For a precise problem formulation we have to specify both the Hough technique used and the assigned goal. We use Hough techniques in which each voting element in the image supplies one vote in the parameter space, either by incorporating edge direction data [3, 22], by using many to one transformations, or by Gerig s method [9] The goal we specify for the Probabilistic Hough Transform is to obtain the same maximum as if the full Hough Transform had been accumulated and analyzed. Possible extensions of the adaptive stopping rules suggested in this paper ....
D. H. Ballard, "Generalizing the Hough Transform to Detect Arbitrary Shapes", Pattern Recognition, Vol. 13, pp. 111-122, 1981.
Guaranteed Convergence of the Hough Transform - Menashe Soffer And (1994) (1 citation) (Correct)
....model is considered and exemplified. Keywords: computer vision, covering methods, global optimization, Hough transform, M estimation, pattern recognition, robust regression 1 Introduction The Hough Transform [7] is a well known technique for detecting collinearities or other predefined patterns [1] in edge images. In this paper the straight line Hough Transform using the normal line parameterization as suggested by Duda and Hart [3] is considered. The strength of the Hough Transform is in fitting straight lines to a collinear subset of a set of points that can include very many outliers, ....
D.H. Ballard, "Generalizing the Hough Transform to Detect Arbitrary Shapes", Pattern Recognit., Vol. 13, pp. 111-122, 1981.
An Optimization Framework for Feature Extraction - Fua And Hanson (Correct)
No context found.
D.H. Ballard, "Generalizing the Hough Transform to Detect Arbitrary Shapes," Pattern Recognition, 13, pp. 111--122 (1981).
Image Informatics At a National Research Center - Rodney Long Sameer (2004) (Correct)
No context found.
Ballard DH. Generalizing the Hough transform to detect arbitrary shapes. Pattern Recogn 1981;13(2):111--22.
Pattern Recognition 32 (1999) 811---824 - Block Decomposition And (Correct)
No context found.
D.H. Ballard, Generalizing the Hough transform to detect arbitrary shapes, Pattern Recognition 13 (1981) 111---122.
Automated Appearance-Based Building Detection In Terrestrial .. - Boehm, Haala, Kapusy (2002) (Correct)
No context found.
D. H. Ballard, 1981. Generalizing the Hough transform to detect arbitrary shapes. Pattern Recognition, 13(2), pp. 111122. E. Baltsavias, A. Grn and L. van Gool, 2001. Automatic Extraction of Man-Made Objects From Aerial and Space Images (III). A.A. Balkema Publishers.
Automatic Segmentation of Cells from Microscopic Imagery.. - Detection Journal Iee (Correct)
No context found.
Ballard, D. H.; "Generalizing the Hough Transform to Detect Arbitrary Shapes", Pattern Recognition, Vol. 13, No. 2, 1981, pp. 111-122.
Parallel Algorithms for Hierarchical Clustering - Clark Olson Computer (1993) (36 citations) (Correct)
No context found.
D. H. Ballard. Generalizing the Hough transform to detect arbitrary shapes. Pattern Recognition, 13(2):111--122, 1981.
Visual Servoing of a Miniature Robot - Toward Marked Target (2002) (Correct)
No context found.
D. Ballard. Generalizing the Hough transform to detect arbitrary shapes. Pattern Recognition, 13(2):111--122, 1981.
Robust Photometric Invariant Features from the Color Tensor - Weijer, Gevers, Smeulders (2004) (Correct)
No context found.
D. H. Ballard. Generalizing the Hough transform to detect arbitrary shapes. Pattern Recognition, 12(2):111--122, 1981. 21
Rectangle Detection based on a Windowed Hough Transform - Jung, Schramm (2004) (Correct)
No context found.
D. Ballard. Generalizing the hough transform to detect arbitrary shapes. Pattern Recognition, 13(2):111--122, 1981.
Invariant Features from Interest Point Groups - Brown, Lowe (2002) (6 citations) (Correct)
No context found.
D. Ballard. Generalizing the Hough Transform to Detect Arbitrary Shapes. Pattern Recognition, 13(2):111--122, 1981.
X-ray Image Analysis - Abeynayake (2001) (Correct)
No context found.
Ballard Dana H. Generalizing the hough transform to detect arbitrary shapes. Pattern Recognition, 13(2):111--122, 1981.
Scale-Invariant Object Categorization using a Scale-Adaptive.. - Leibe, Schiele (2004) (1 citation) (Correct)
No context found.
D.H. Ballard. Generalizing the hough transform to detect arbitrary shapes. Pattern Recognition, 13(2):111--122, 1981.
The Robutler: a Vision-Controlled Hand-Arm System.. - Hillenbrand.. (2004) (Correct)
No context found.
D. H. Ballard, "Generalizing the Hough transform to detect arbitrary shapes," Pattern Recognition, vol. 13, pp. 111--122, 1981.
Environment Learning For Indoor Mobile Robots - Cetto (2003) (Correct)
No context found.
D. H. BALLARD, Generalizing the hough transform to detect arbitrary shapes. Pattern Recognition, vol. 13, no. 2, pp. 111--122, 1981.
Performance Evaluation of Object Recognition Techniques - Ulrich, Steger (2002) (Correct)
No context found.
Ballard, D. H., 1981. Generalizing the Hough transform to detect arbitrary shapes. Pattern Recognition 13(2), pp. 111--122.
Combined Object Categorization and Segmentation With An.. - Leibe, Leonardis.. (2004) (1 citation) (Correct)
No context found.
D.H. Ballard. Generalizing the hough transform to detect arbitrary shapes. Pattern Recognition, 13(2):111--122, 1981.
Accurate Estimates Of False Alarm Number In Shape Recognition - Muse, al. (2004) (Correct)
No context found.
D.H. Ballard. Generalizing the Hough transform to detect arbitrary shapes. Pattern Recognition, 13(2):111 122, 1981.
The Robutler: a Vision-Controlled Hand-Arm System.. - Hillenbrand.. (Correct)
No context found.
D. H. Ballard, "Generalizing the Hough transform to detect arbitrary shapes," Pattern Recognit., vol. 13, pp. 111--122, 1981.
An Optimization Framework for Feature Extraction - Fua And Hanson (1991) (9 citations) (Correct)
No context found.
D.H. Ballard, "Generalizing the Hough Transform to Detect Arbitrary Shapes," Pattern Recognition, 13, pp. 111--122 (1981).
Interleaved Object Categorization and Segmentation - Leibe, Schiele (2003) (3 citations) (Correct)
No context found.
D.H. Ballard. Generalizing the hough transform to detect arbitrary shapes. Pattern Recognition, 13(2):111--122, 1981.
A short introduction to the Radon and Hough transforms and .. - van Ginkel, van Vliet (2004) (Correct)
No context found.
D.H. Ballard. Generalizing the Hough transform to detect arbitrary shapes. Pattern Recognition, 13(2):111--122, 1981.
A New Fast and Robust Circle Extraction Algorithm - Euijin Kim Miki (2002) (Correct)
No context found.
D. H. Ballard, "Generalizing the Hough transform to detect arbitrary shapes", Pattern Recognition, vol.13, no.2, pp.111--122, 1981.
Robust Curve Detection using a Radon Transform in.. - van Ginkel.. (2003) (Correct)
No context found.
D.H. Ballard. Generalizing the Hough transform to detect arbitrary shapes. Pattern Recognition, 13(2):111--122, 1981.
Genetic Object Recognition Using Combinations of - Views George Bebis (2002) (Correct)
No context found.
D. Ballard, "Generalizing the hough transform to detect arbitrary patterns, " Pattern Recognit., vol. 13, no. 2, pp. 111--122, 1981.
Linear Color Segmentation and its Implementation - Dmitry Nikolaev And (Correct)
No context found.
D. Ballard, Generalizing the Hough Transform to detect arbitrary shapes, Pattern Recognit. 13, 1981, 111-122.
Dressed Human Modeling, Detection, and Parts Localization - Zhao (2001) (Correct)
No context found.
D.D. Ballard, "Generalizing the Hough Transform to Detect Arbitrary Shapes," Pattern Recognition, Vol. 13, No. 2, pp. 111-122, 1981.
Detection of Unknown Forms from Document Images - Busch, Boles, Sridharan.. (2003) (Correct)
No context found.
D. H. Ballard, "Generalizing the Hough transform to detect arbitrary shapes," Pattern Recognition, vol. 13, pp. 111-122, 1981. 144
Unsupervised Thresholds For Shape Matching - Pablo Mus Fr (2003) (Correct)
No context found.
D.H. Ballard, "Generalizing the Hough transform to detect arbitrary shapes," Pattern Recognition, vol. 13, no. 2, pp. 111--122, 1981.
The Hough Transform As A Tool For Image - Analysis Josep Llados (Correct)
No context found.
D.H. Ballard. Generalizing the Hough transform to detect arbitrary shapes. Pattern Recognition, 13(2):111--122, 1981.
Dressed Human Modeling, Detection, and Parts Localization - Zhao (2001) (Correct)
No context found.
D.D. Ballard, "Generalizing the Hough Transform to Detect Arbitrary Shapes," Pattern Recognition, Vol. 13, No. 2, pp. 111-122, 1981.
Affine Matching With Bounded Sensor Error: A Study of.. - Grimson, al. (1991) (13 citations) (Correct)
No context found.
Ballard, D.H., 1981, "Generalizing the Hough Transform to Detect Arbitrary Patterns," Pattern Recognition 13(2): 111-122.
Appendix - Projective Geometry for Machine Vision - Mundy, Zisserman (1992) (7 citations) (Correct)
No context found.
Ballard, D.H., Generalizing the Hough Transform to Detect Arbitrary Shapes, Pattern Recognition, Vol. 13, No. 2, p.111-122, 1981.
General-Purpose Acousto-Optic Connectionist Processor - Naughton, Javadpour, Keating (1999) (Correct)
No context found.
D. H. Ballard, "Generalizing the Hough transform to detect arbitrary shapes," Pattern Recog. 13~2!, 111--122 ~1981!.
First 50 documents Next 50
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC