J. Burns, A. Hanson, and E. Riseman, "Extracting Straight Lines," pp. 425-455, July 1986.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....under different lighting conditions . Image size: 400x400 pixels . Detection rate: 100 . Positional accuracy of circle: 1 pixel. Figure 5. 18 Detected circles of a watermeter Rectangle detection A rectangle consists of 4 straight lines, which we detect with the approach of Burns et al. [BUR86]. This algorithm groups pixels into line support regions of similar gradient orientation. The structure of the associated intensity surface determines the location and properties of the edge. The line extraction is more effective than previous techniques (for example Hough transform [HOU62] ....

J.B. Burns, A.R. Hanson, E.M. Riseman, "Extracting Straight Lines", IEEE Trans. Pattern Anal. Mach. Intell., Vol.8, No.4, pp.425-455, 1986.

...., proximity, and similarity are the fou Gestalt principles of perceptuP organization that have been appliedquli suiedgbzbb1 infeatuM detection and sceneunegb6bg bM machine vision. Linkingand merginga set of detected edge elements into more prominentfeatuen sut as line and e segments [3] is a typical application ofperceptuz organization. Distingun.g ingfigu1 fromgrou1 is another basic and powerfu Gestalt A.H.H.Ngu et al. Combiningmuiningg1MM featuin for efficient indexing in a large image database principle ofvisuP perceptuk organization. When we are presented with an ....

J.B. Bu.gz A.R. Hanson, E.M. Riseman (1984) Extracting straight lines. In: Int. Conf. on Pattern Recognition 1:482--485 4. T.ChiuM (1994) Content-based image indexing. In: VLDB'94,

....problem in under 20 minutes. We have also solved other large real world geometric matching problems using line segments extracted from photographs with up to 20,000 model line data line segment pairs. a) b) c) d) Figure 4: Real data example. a) aerial photograph, b) Burns line segments [Burns et al. 1986], c) building model, d) best match. 3 Fast Messy Genetic Algorithms For the geometric matching problem we used a customized version of the Messy Genetic Algorithm; we now look at a relatively generic version of the MGA and apply it to a set of synthetic test problems. The Fast Messy Genetic ....

Burns, J. B., Hanson, A. R., and Riseman, E. M. (1986). Extracting straight lines. IEEE Trans. on Pattern Analysis and Machine Intelligence, PAMI--8(4):425 -- 456.

....constraints generated from the caption of Figure 1. and the extraction of visual properties. The face location module [Govindaraju et al. 1992] is an object locator which uses edge contours as basic features, and a springs and templates model in a multi resolution framework. A line detector [Burns et al. 1984] is being used to construct simple detectors for 2D rectilinear objects. Figure 5.1 is an example of a digitized newspaper photograph and the accompanying caption. The task is to correctly identify each of the seven people mentioned in the caption. Information from the caption (in the form of a ....

J. Burns, A. Hanson, and E. Riseman. Extracting straight lines. Technical Report 84-29, University of Massachusetts, Amherst, 1984.

....process which groups primary edge pixels into line structures. Hough techniques can directly extract straight lines, but the information about the end points of a line segment and it grey attribute is not provided. Hence they are not very suited for line based stereo. For this goal, Burns s method [2] seems to be more attractive. This method firstly segments neighboring pixels with similar gradient directions into line support regions and then directly extracts straight lines by fitting planes, where step like edges are assumed. When applying it to a real image it is difficult to get the ....

J.B.Burns, A.R.Hanson and E.M.Riseman, Extracting straight lines, IEEE Trans. Pattern Anal. Mach. Intelligence, Vol. 8, pp. 425-455, 1986.

....unlikely to form meaningful structures. Manmade objects usually consist of regular structures related structurally and spatially. The subsequent sections describe in detail the extraction of each of these data items. 2. 4 Extraction of straight line segments Burns straight lines detector [43] has been used to extract straight line segments. The distinction between the terms line segments and lines will be made on the basis that a line is obtained by grouping these fragmented line segments. Burns operator uses the orientation of the local gradient in the neighborhood of a pixel to ....

J. Brian Burns, Allen R. Hanson, and Edward M. Riseman, "Extracting straight lines," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 8, no. 4, pp. 425--455, 1986.

....implicitly use an intensity model [6] 3] Masks are used as templates: the best fit provides the operator output and the characteristics of the template are assigned to the pixel. The results thus depend on how well the templates may represent a given line. Other detectors use the gradient; in [1] for example, a segmentation based on the gradient orientation guides the road detection. Finally, some road finders use the edges as in [7] where the author defines road centre hypotheses as anti parallel intensity edges. In this paper, we show that imposing constraints on the gradient field ....

....(1D) line is shown on Figure 1. If the line is lighter than the background, the derivative of the intensity will be positive at the left side of the bar and negative at the right side. As data are discrete, the derivative is computed by convolving the signal with a kernel; for example, the kernels [ 1 0 1] and [ 0.04 0.28 0.56 0 0.56 0.28 0.04] respectively provide the discrete derivative and the derivative of the intensity regularized by a Gaussian. Let f 0 (i) denote the derivative at pixel i. Then Gammaf 0 (i Gamma 1) Theta f 0 (i 1) will be null outside the bar, negative around the ....

J.B. Burns, A.R. Hanson, and E.M. Riseman, "Extracting straight lines", IEEE Trans. on Pattern Anal. and Machine Intelligence PAMI-8:425--455, Jul. 1986.

....properties. The face location module [ Govindaraju et al. 1992 ] is an object locator which uses edge contours as basic features, and a springs and templates model in a multi resolution framework. We are currently improving a simple neural network based gender discriminator. A line detector [ Burns et al. 1984 ] is being used to construct simple detectors for 2D rectilinear objects. Figure 1 is an example of a digitized newspaper photograph and the accompanying caption. The task is to correctly identify each of the seven people mentioned in the caption. Information from the caption (in the form of a ....

J. Burns, A. Hanson, and E. Riseman. Extracting straight lines. Technical Report 8429, University of Massachusetts, Amherst, 1984.

....as scale invariance. A new method is described that adheres the formulated criteria. Finally an experiment is presented that illustrates the usefulness of the new method. 1 Introduction In computer vision, the detection of lines is an important and longstanding topic. Many authors, for example [Burns86], have described methods for detecting straight lines in grey value images. In this paper we consider the problem of clustering straight line segments into new ones. The clustering of line segments is set up as a hierarchy. The hierarchy gives a priority on what original line segments are ....

J.B. Burns. Extracting Straight Lines. IEEE PAMI 8(4), pp. 425-455, 1986.

....module is to take the edges that are detected and organize them into more perceptually meaningful structures. This can be done by grouping neighboring image tokens, as perceptual organization would suggest. Much of the previous research in this area has been dedicated to detecting straight lines [53, 32, 54] or curves whose model is known a priori [55, 56] More general algorithms have also been studied by Lowe, whose algorithm emphasized recursively grouping adjacent edge points into lines at multiple scales [14] But he used a linear or circular curve model for performing the perceptual ....

J. B. Burns, A. R. Hanson, and E. M. Riseman. Extracting straight lines. In M. A. Fischler and O. Firschein, editors, Readings in Computer Vision, Issues, Problems, Principles, and Paradigms, pages 180-183. Morgan Kaufmann, 1987.

....with robust output. Applications for a real time line extraction algorithm at UMass are the Unmanned Ground Vehicle (UGV) and the algorithms implemented on it. The line extraction algorithm was developed, with the goal to be as fast as e.g. the FLF [4] and as accurate as e.g. the Burns algorithm [1]. It is based on the line extraction algorithm of Burns, but was modified significantly. Developing the algorithm on the IUA, some design criteria in the algorithm were made considering the architecture of the IUA. 4.1 The Burns Algorithm First, some basic principles of the Burns algorithm are ....

....for the multiply and assignment plus the setting of the activity. The term time to build Select1 indicates the time to set the activity, where one processor is active at each row. This is also the activity which is set for the functions with suffix with[in] Select. The term time to build Coterie[1] indicates the time to set up the Coterie network, where the Coterie networks correspond to the suffix of the above introduced Coterie functions. The basic functions give some first insight into the time ranges on the three machines. For functions that are missing above, similar functions ....

[Article contains additional citation context not shown here]

J.B. Burns, A.R. Hanson and E.M. Riseman. "Extracting Straight Lines." IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. PAMI-8, No. 4, July 1986.

....developed to evaluate complete image understanding systems [24] Uses the low level bottom up and the intermediate level processing components. Weymouth Overton Preprocessor Information preserving image filter. Uses edge model curve fitting [14] Fast Line Finder Based on Burns algorithm [3]: segments image by gradient orientation and fits line to resulting regions. Depth From Motion Computation dominated by correspondence based matching [7] Boldt s Line Finder Perceptual organization based. Iterative grouping algorithm [2] Region Segmentation Based on Nagin Kohler system [1] ....

J.B. Burns, A.R. Hanson, E.M. Riseman (1986): "Extracting Straight Lines," IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI-8 (4) pp. 425-455.

....to parts of the scene being translated outside the camera s field of view. The jitter is replaced by whatever positional inaccuracies come from an actual camera producing an 128 Theta 128 image (Williams and Hanson, 1988) which is then processed by a high quality line segment finding algorithm (Burns, 1986). Better results would be expected of objective functions derived from grammars which explicitly model more of these noise processes, such as the grammars studied in Section 4. Since the data to be matched are not dots but line segments, one could alter the grammar and rederive the various ....

Burns, J. B. (1986). Extracting straight lines. IEEE Trans. PAMI, 8(4):425--455.

....the accumulated errors[7] Fig. 11 summarizes the procedure of localization performed in this research. A single frame image from the mono vision system is processed to extract linear features in indoor environment such as edges along the floor and doors. Canny s operator[8] and Hough transform[9] are modified and utilized for edge detection and line detection, respectively. The positions of the detected linear features are inversely mapped from the image coordinate to the world coordinate by 1 A . Then, the mobot s actual position is computed by referencing the positions of door ....

B. Burns, et. al., "Extracting Straight Lines," IEEE Trans. on on Pattern Analysis and Machine Intelligence, Vol. 8, No. 4,

.... to multiresolution approaches by Cook and Delp [6, 7] Other approaches have also been reported in the literature [8, 9, 10] The use of local edge operators to first enhance the image usually also enhances noise and tends to generate dense edge maps, which makes subsequent processing di#cult [11]. Furthermore, these methods do not respond well to thick lines because pixels in the middle of a thick line are not edge pixels. We will show examples of these problems in Section 2. Another approach is to parameterize potential lines in edges detected in the scene and then base the detection in ....

J. B. Burns, A. R. Hanson, and E. M. Riseman, "Extracting straight lines," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 8, no. 4, pp. 425--455, July 1986.

.... three options are: 1) an active contour procedure [18] that modifies the outline of an ROI mask to maximize the energy under the boundary; 2) a segmentation procedure [10] that extracts the boundary of a new region (as a 2D contour) within the image chip; or 3) a straight line extraction procedure [7]. A Generalized Hough Transform procedure [5] matches 2D image contours to the contour of a template, thus creating a new ROI. A symbolic line matching procedure (LiME; 6] finds the rotation, translation and scale that maps template (model) lines onto image lines, again producing an ROI. It ....

B. Burns, A. Hanson and E. Riseman. "Extracting Straight Lines," PAMI 8(4):425-455, 1986.

....process which groups primary edge pixels into line structures. Hough techniques can directly extract straight lines, but the information about the end points of a line segment and it grey attribute is not provided. Hence they are not very suited for line based stereo. For this goal, Burns s method [2] seems to be more attractive. This method firstly segments neighboring pixels with similar gradient directions into line support regions and then directly extracts straight lines by fitting planes, where step like edges are assumed. When applying it to a real image it is difficult to get the ....

J.B.Burns, A.R.Hanson and E.M.Riseman, Extracting straight lines, IEEE Trans. Pattern Anal. Mach. Intelligence, Vol. 8, pp. 425-455, 1986.

....constraints generated from the caption of Figure 1. and the extraction of visual properties. The face location module [Govindaraju et al. 1992] is an object locator which uses edge contours as basic features, and a springs and templates model in a multi resolution framework. A line detector [Burns et al. 1984] is being used to construct simple detectors for 2D rectilinear objects. Figure 5.1 is an example of a digitized newspaper photograph and the accompanying caption. The task is to correctly identify each of the seven people mentioned in the caption. Information from the caption (in the form of a ....

J. Burns, A. Hanson, and E. Riseman. Extracting straight lines. Technical Report 84-29, University of Massachusetts, Amherst, 1984.

....Genetic Algorithm runs 9:4 times faster. In other words, for the problems taking thousands of seconds to solve using local search, the Messy Genetic Algorithm is dropping run times by an order of magnitude. a) b) c) d) Figure 4: Real data example. a) aerial photograph, b) Burns line segments [Burns et al. 1986], c) building model, d) best match. Figure 4 illustrates a matching problem involving data from a real image. For this problem, there are 4 model line segments, 443 data line segments, generating 1; 772 possible pairs of segments. This problem is hard both because the search space is large, 2 ....

Burns, J. B., Hanson, A. R., and Riseman, E. M. (1986). Extracting straight lines. IEEE Trans. on Pattern Analysis and Machine Intelligence, PAMI--8(4):425 -- 456.

.... the Canny edge detector [Can86] The edge detector convolves the image with the derivative of a gaussian filter, and searches for maxima of gradient magnitude (along the gradient direction) Edge line segments are identified by connecting adjacent pixels that possess the same gradient direction [BHR87] Corners are determined by intersecting adjacent line segments, i.e. whose end points are within a threshold distance (e.g. one pixel) Figure 1 6 shows feature detection performed on a single image. The features detected are stored along with their corresponding images, so that they can be ....

J.B. Burns, A.R. Hanson, and E.M. Riseman. Extracting straight lines. In Readings in Computer Vision, pages 180--183, 1987.

....from the image until two vectors pointing from the (variable) focal point towards two (fixed) vanishing points in the image are perpendicular. 4. 1 Model to Image Matching Figures 1a) and b) show a set of straight line segments extracted from an image of a wall poster using the Burns algorithm [10], and a set of model lines to be matched to the image. The first stage in matching is to detect two clusters of lines converging to the two main vanishing points in the image, and from the resulting vanishing line rectify the image to present a frontal view of the poster (Figure 1c) The four ....

J.B. Burns, A.R. Hanson and E.M. Riseman, "Extracting Straight Lines," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 8, No. 4, July 1986, pp.425-456.

....verification phase. The method also has trouble finding short segments in busy images. Princen et al. 13] address some of these problems using a hierarchical grouping process in conjunction with a local Hough transform. A third method of lineal feature detection due to Burns et al. [1] utilizes the gradient direction to partition the image into a set of support regions, each of which will presumably be associated with a single feature. A least squares fitting procedure is then used to fit a line segment to each region. This method can detect low contrast features, but the ....

J. B. Burns, A. R. Hanson, and E. M. Riseman. Extracting Straight Lines. In Proc. DARPA IU Workshop, pages 165--168, New Orleans, LA, 1984.

.... 1) an active contour procedure [18] that modifies the outline of an ROI mask until the contour lies along edges in the original image; 2) a segmentation procedure [10] that extracts the boundary of a new region (as a 2D contour) within the image chip; or 3) a straight line extraction procedure [7]. A Generalized Hough Transform procedure [5] matches 2D image contours to the contour of a template, thus creating a new ROI. A symbolic line matching procedure (LiME; 6] finds the rotation, translation and scale that maps template (model) lines onto image lines, again producing an ROI. It ....

B. Burns, A. Hanson and E. Riseman. "Extracting Straight Lines," PAMI 8(4):425-455, 1986.

....of two line detection algorithms. The task was to detect the presence or absence of a vertical edge in the middle of an image containing a grating mask and additive Gaussian noise. We compare two line detection algorithms: i) the facet edge detector [11] followed by the Burns line finder [12] and (ii) the facet edge detector followed by the Hough transform. The methodology is defined in a general enough way that it can be readily applied to other computer vision algorithms that can be specified as a detection task. Typical problems where this methodology can be applied are the ....

....stage labels individual pixels as edge or non edge. The line detection stage identifies groups of edge pixels that might represent a line. For edge detection we use the second directional derivative facet edge detector [11] For the second stage, the line detection is done by the Burns line finder [12]. The Burns line finder takes the output of the edge detector and labels all the connected lines it finds. As a measure of the confidence of the algorithm in the presence (or absence) of a vertical line at the location of interest (the column of pixels at or near the center of the image) we ....

J.B. Burns, A.R. Hanson, and E.M. Riseman, "Extracting straight lines", IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 8, pp. 425--455, July 1986.

....in which holes for the ground plug either were or were not present. The recognition strategy is based on the idea of repeatedly applying grouping and filtering operations. First, it attempts to find the boundaries of the socket plate. Straight lines are extracted using the algorithm described in [Burns et al. 1986] with the speedups described in [Kahn et al. 1990] Since sockets are most likely to be mounted on a vertical 10 In fact, sockets with things plugged in are an excellent example of a task that a conventional geometric model based system would find quite difficult, since plugs come in so many ....

J.B. Burns, A.R. Hanson, and E.M. Riseman, "Extracting Straight Lines," IEEE Transactions on Pattern Analysis and Machine Intelligence, 8:425--455, July 1986.

.... most relevant tokens are analyzed first by the recognition process [4] The system currently uses primitives provided by three different types of segmentation processes: line segments, circular arcs and uniform regions (Figure 5) The line segment extraction is performed using a standard algorithm [7]. A filter is then applied to remove line segments shorter than a certain threshold (Figure 5.a) Circular arcs are obtained from a least squares fit to the chains provided by the Canny edge detection algorithm [8] In addition to the least squares approximation, the fitting algorithm has an ....

J.B. Burns, A.R. Hanson and E.D. Riseman, "Extracting Straight Lines", IEEE Trans. on PAMI, vol 8(4), 1986, pp. 425-455.

....Hough transform [2] Here local edges vote for all possible lines they are consistent with, and the votes are summed up later to determine what lines are actually present. The disadvantages of this method are complexity, coarse resolution and lack of locality. The method developed by Burns et al. [1] partitions the image into a set of support regions based on the gradient direction. Each region will presumably be associated with a single feature. A method developed by Nelson [11] uses both gradient magnitude and direction information and incorporates explicit lineal and end stop terms. These ....

J. B. Burns, A. R. Hanson, and E. M. Riseman. Extracting straight lines. In PROC. DARPA IU Workshop, pages 165--168, New Orleans, LA, 1984.

....our clustering method using hand written drawings and obtained promising results. 1 Introduction Extraction of lines from an image is an essential task in computer vision. Many authors have reported various algorithms for line extraction, such as Hough Transformation and gradientbased methods[2]. An effective approach to line extraction is by clustering line segments [5, 6] Line segments are obtained by applying a line fitting process to the output of edge detection process. A line segment is a small geometric structure defined by two end points. In many cases, line segments lie along ....

Burns, J. B., A. R. Hanson and E. M. Riseman (1986). Extracting Straight Lines. IEEE Trans. Patt. Anal. Mach. Intell. 8 (4), 425-- 455.

....information extracted from the input image that is later used for recognition. In particular, three types of image primitives, also called simple tokens , are extracted: line segments, circular arcs and uniform regions (Figure 6) The line segment extraction is performed using a standard algorithm [Burns86]. A filter is then applied to remove line segments shorter than a certain threshold (Figure 6.a) Circular arcs are obtained from a least squares fit to the chains provided by the Canny edge detection algorithm (Figure 6.b) Canny86] The region segmentation algorithm is based on two separate ....

J.B. Burns, A.R. Hanson, and E.D. Riseman, "Extracting Straight Lines", IEEE Trans. on PAMI, 8(4), 1986, 425-255.

No context found.

Burns, J. B., Hanson, A. R., and Riseman, E. M., Extracting straight lines. IEEE Trans. PAMI, Vol. 8(4), July 1986.

....elevation. 18 plane finder from generating regions that cross these lines, thus improving the accuracy of the planes near edges. Figure 22 shows an example result of this procedure applied to the ISPRS data. First, straight lines were extracted from one of the two optical stereo images using Burns (1986) algorithm. Then, these lines were removed from the elevation map, and the plane finder was run with the same parameters as used in Fig. 15c. The improvement is striking: Without the line extraction, the plane finder had difficulties precisely locating the region boundaries because of the absence ....

Burns, J. B., A. R. Hanson, and E. M. Riseman (1986). Extracting straight lines. IEEE Transactions on Pattern Analysis and Machine Intelligence 8 (4), 425--455.

.... examples are hand built, local search has been used with real building models on the RADIUS calibrated terrain board imagery [18] It has also been used with aerial photgraphs to register ortho rectified images [7] The data line segments for these examples are produced using the Burns algorithm [19]. For the match in Figure 2d, the model is rotated by 120 ffi , illustrating that the original orientation of the model does not matter. In this example, there is little clutter and the building has a distinctive form. We ll see in Section VII that his is not a difficult problem. The match in ....

J. B. Burns, A. R. Hanson, and E. M. Riseman, "Extracting straight lines," IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. PAMI--8, no. 4, pp. 425 -- 456, July 1986.

....coordinates into observed image plane coordinates is Cam(u 0 ; v 0 ; f) 2 6 4 f 0 u 0 0 f v 0 0 0 1 3 7 5 : 4.10) 115 Computing Vanishing Point Orientations Figure 4.6 shows a set of line segments produced from the image in Figure 4. 5 by the Burns straight line extraction algorithm [Burns86], as implemented in the KBVision image understanding development system [AAI91] Line extraction was run on each of the eight images J1 J8. Due to the large size of these images, they were first reduced in resolution to half their original size before line extraction. After line extraction, the ....

....and a contrast (difference in average grey level across the line) of at least 15 grey levels. This procedure produced roughly 5000 line segments per image. Figure 4.6: Line segments extracted for Radius model board image J8. These lines were found using the Burns straight line extraction algorithm [Burns86]. 116 Theoretically, both vanishing point detection and estimation are invariant to the camera calibration parameters. To test this is practice, both vanishing point detection and estimation were carried out using the following default camera model u 0 = ImageWidth = 2 (4.11) v 0 = ....

J Burns, J.B., Hanson, A.R. and Riseman, E.M. "Extracting Straight Lines," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 8, No. 4, July 1986, pp. 425456.

No context found.

J. Burns, A. Hanson, and E. Riseman, "Extracting Straight Lines," pp. 425-455, July 1986.

No context found.

J. Burns, A. Hanson, and E. Riseman, "Extracting straight lines," IEEE Trans. on PAMI, vol. 8, no. 4, pp. 425--455, July 1986.

No context found.

J. Burns, A. Hanson, E. Riseman, Extracting straight lines, IEEE Trans. PAMI 8 (4) (1986) 425--455.

No context found.

J. B. Bums, A. R. Hanson, and E. M. Riseman, "Extracting straight lines," IEEE Trans. Pattern Anal. Machine Intell., vol. PAMI-8, pp. 425--455, July 1986.

No context found.

J. B. Burns, A. R. Hanson, and E. M. Riseman, "Extracting straight lines," IEEE Trans. Pattern Anal. Machine Intell., vol. 8, pp. 425--455, July 1986.

No context found.

J. Burns, A. Hanson, and E. Riseman, "Extracting straight lines," IEEE Trans. on PAMI, vol. 8, no. 4, pp. 425--455, July 1986.

No context found.

J.B. Burns, A.R. Hanson, E.M. Riseman, "Extracting Straight Lines", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. PAMI-8, No.4, pp.425-455, 1986

No context found.

Burns, JB, Hanson, AR, Riseman, EM. Extracting straight lines. University of Massachusetts COINS Technical Report 84-29, December 1984. References not used:

No context found.

Burns, JB, Hanson, AR, Riseman, EM. Extracting straight lines. University of Massachusetts COINS Technical Report 84-29, December 1984.

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