Showing papers on "Canny edge detector published in 1990"

On the localization performance measure and optimal edge detection

Abstract: The localization performance measure of edge detectors is addressed. A one-dimensional formulation of the problem is considered. A linear space-invariant filter is used for the detection. The locations of local maxima in the thresholded output of the filter are declared to be the edges. The limitations of conventional performance measures are shown, and a localization performance measure for edge detection is suggested. This performance measure is based on the theory of zero-crossings of stochastic processes. It is shown that the derivative of a Gaussian is the optimal edge detector for the measure. >

Recursive Filtering and Edge Closing: two primary tools for 3-D edge detection

Abstract: This paper deals with edge detection in 3D images such as scanner, magnetic resonance (NMR), or spatio-temporal data. We propose an unified formalism for 3D edge detection using optimal, recursive and separable filters recently introduced for 2D edge detection. Then we obtain some efficient 3D edge detection algorithms having a low computational cost. We also show that 3D edge tracking/closing enables to extract many edges not provided by the filtering stage without introducing noisy edges. Experimental results obtained on NMR images are shown.

Edge detection using a neural network

Abstract: An edge detection algorithm using multistate ADALINES (adaptive linear neurons) is presented. The proposed algorithm can suppress noise effects without increasing the mask size. The input states are defined using the local mean in a predefined mask, and the one-dimensional edges are defined so that they are linearly separable from nonedges. The two-dimensional edges are obtained using the rotation invariant property of layered neural networks. The proposed algorithm requires much less computation compared with Marr and Hildreth's (1980) edge detector for similar performance. An application of the proposed edge detector to adaptive image restoration is also presented. >

Optimal filter for edge detection methods and results

Abstract: In this paper, we give a new demonstration in which it is proved that the symmetric exponential filter is the optimal edge detection filter in the criteria of the signal to noise ratio, localization precision and unique maximum. Then we deduce the first and the second directional derivative operators for symmetric Exponential Filter and realize them by first order recursive algorithm, and propose to detect the edges by maxima of Gradient (GEF), or by the zeros crossing of Second directional Derivative along the gradient direction (SDEF).

Smoothed differentiation filters for images

Abstract: A systematic approach to least square approximation of images and of their derivatives is presented. Derivatives of any order can be obtained by convolving the image with a priori known filters. It is shown that if orthonormal polynomial bases are employed the filters have closed-form solutions. The same filter is obtained when the fitted polynomial functions have one consecutive degree. Moment-preserving properties, sparse structure for some of the filters, and the relationship to the Marr-Hildreth and Canny edge detectors are proven. >

A model-based approach for filtering and edge detection in noisy images

Abstract: The authors consider the problem of enhancement and edge detection on noisy, real-world images. The restoration and edge detection framework is based on an autoregressive (AR) random-field model. An edge is detected if the first and second directional derivatives and a local estimate of the variance at each point satisfy certain criteria. When noise is present, a good estimate of the original from the noisy images improves the signal-to-noise ratio, resulting in better estimates of the directional derivatives. To avoid excessive computation, the problem of estimation of the original image and the model parameters is presented as a combination of a reduced-update Kalman filter and an adaptive-least-squares parameter estimation algorithm. The restoration process is completed with a min-max replacement scheme to enhance edge strength. An orientation-sensitive detector resulting from the use of an AR model may not detect edges of significantly different orientations. This is partially overcome by running four edge detectors on the four interior pixels of a 4*4 window; this corresponds to rotating the window in successive multiples of 90 degrees . Comparisons with R.M. Haralick's (1984) facet model edge detector, R. Nevatia and K.R. Babu's (1980) line finder, and J. Canny's (1986) edge detector are given. >

Statistical Edge Detection Operators for Linear Feature Extraction in SAR Images

Abstract: Edge detection in synthetic aperture radar (SAR) images is rendered difficult by the presence of speckle. The data are often filtered using adaptive filters independently of the edge detection process when, in fact, the two steps should be coupled (i.e., the local homogeneity criterion employed by an adaptive filter should be consistent with the edge detector criterion). Five different edge detection algorithms for SAR images are evaluated and compared. The detection algorithms are comprised of two operators based on non-parametric statistical tests, the Ratio of Averages test, difference of averages (essentially a gradient method), and a test based on the mean squared to variance ratio. Two edge thinning and thresholding operations are also compared: an algorithm proposed by Nevatia and Babu (1980), and one based on mathematical morphology (Serra, 1980). Initial testing is carried out on simulated imagery for accurate control of the signal being masked by speckle noise. We obtain the best results using t...

3D edge detection by separable recursive filtering and edge closing

Abstract: Edge detection in 3D images such as scanner, magnetic resonance, or spatiotemporal data is considered. A two-stage scheme based on separable recursive filtering and edge tracking/closing is proposed. The key point of the filtering stage is to use optimal recursive and separable filters to approximate gradient or Laplacian methods. The recursive nature of the operators enables one to implement infinite 3D impulse response with a computing time roughly similar to a 3*3*3 convolution mask. The principle of the edge tracking/closing is to select from the previous stage only the more reliable edge points and then to apply an edge closing method derived from the idea developed by R. Deriche and J.P. Cocquerez (1988). This makes it possible to substantially improve the results provided by the filtering stage. >

On local detection of moving edges

Abstract: The authors propose a detection framework with multiple velocity channels for moving edges based on a generalization of J. Canny's edge detector (1986). Finite state machines (FSMs) are set up at discrete lattice points in the image plane and operate based on the outputs of all velocity channels. The outputs of the FSMs denote whether there are edges at their corresponding positions, and their states record the edge velocities. In the temporal dimension, statistics are attached to the edges to aid in removing phantom edges. >

Unification of image segmentation and edge detection

Abstract: A method of simultaneous image segmentation and edge detection based on grey-level co-occurrence matrices is described. An analysis of the distributions within a co-occurrence matrix defines an initial pixel classification into both region and interior or boundary classes. Local consistency of pixel classification is enforced by minimising the entropy of local region and boundary information, where region information is expressed by conditional probabilities, estimated from the co-occurrence matrices, and boundary information by conditional probabilities which are determined a priori. The method robustly segments an image into homogeneous areas and generates an edge map. The technique extends easily to general edge operators; examples are given of the techniques applied to both synthetic and infrared imagery for the [1 –1] and Canny edge operators. The results are compared with other techniques.

A spatio-temporal generalization of Canny's edge detector

Abstract: Moving step edges are modeled as the product of a deterministic function in space and a stochastic function in time which captures the edge shapes and the temporal uncertainties, respectively Under J Canny's (IEEE Trans on Pattern Analysis and Machine Intelligence, volPAMI-8, p679-98, Nov 1986) original optimality criteria, a set of optimal edge detectors is derived They are in a product form, ie, a product of a spatial function and a temporal function The spatial function is Canny's edge detector in one dimension and the temporal function can be well approximated by the exponential function Generalizing Canny's edge detector to the temporal domain is not only theoretically interesting, but also practically useful The generalization of Canny's edge detectors provides better immunity to noise and can serve as one of the tools in understanding the temporal behavior of moving edges They have been used in a data-fusion framework to detect moving edges and their normal velocities simultaneously For completeness, the authors derive some properties of the optimal edge detectors and compare them with Gabor filters >

Detection and smoothing of edge contours in images by one-dimensional Kalman techniques

Abstract: An edge model is developed along with the associated edge detection and contour determination algorithms. An edge point along any scan direction is defined as a sufficient jump in the mean of a time series that is assumed to be white Gaussian around the edge point with the same variance on either side. An edge contour is defined as a sequence of such edge points, each point denoted by the two coordinates of the image plane. Each coordinate sequence is modeled as first-order autoregressive Gaussian over and above a straight line sequence of arbitrary, finite slope and intercept. The first stage of the overall algorithm examines the time series along each row and each column, forms a window around each potential edge point, detects the location and estimates the variance of the error in the location of the edge point, by pattern recognition techniques. The second stage forms noisy edge contours by graph searching techniques. The third stage is smoothing of an edge contour, formulated as a well-defined Kalman smoothing problem. It is shown that the parameters of the system and noise can all be estimated from the data itself. Experimental results on a simple image are discussed. >

Fast Implementation Of A Laplacian Of Gaussian Edge Detector

Abstract: The zero-crossing test of the second directional derivative is regarded by many image processing researchers as the optimal method of edge detection. Fast implementation of a zero-crossing detector for an image that has been operated upon by a Laplacian of Gaussian (LOG) convolution is difficult; execution speed is, in fact, the primary disadvantage of edge detection by LOG. Another drawback to this method is the difficulty in programming the algorithm itself. Use of efficient library functions to perform the required LOG convolutions and tradeoffs between multiple cascaded filters and a single filter with a spatially large impulse response will be examined. Finally, a convolution and look-up table-based implementation of a zero-crossing detector will be explained.

Localization performance measure and optimal edge detection

Abstract: Recent developments in edge detection have exposed diiferent criteria to gauge the performance of edge detectors in the presense of noise. One of the criteria is "Localization", which is the ability of the edge detector to produce from noisy data a detected edge that is as close as possible to the true edge in the image. In this paper, we show the limitation of the localization criteria as previously formulated and propose an alternative. This new performance measure is based on the theory of zero-crossings of stochastic processes. We show that the derivative of a Gaussian is the optimal edge detector for this new measure.

A circular smoothing edge detector

Abstract: A method of detecting step edges based on an edge-preserving circular smoothing technique is presented. The detection procedure consists of two stages: circular smoothing and thresholding. To achieve better results, multiple windows are used. Experimental results utilizing real images show that using multiple windows can more efficiently remove noise, reduce edge width, and detect weak edges. Based on the edge quality, the method is comparable to the commonly used edge detectors, but it requires fewer steps. Since the edge detector can be implemented in parallel, it is suitable for VLSI implementations leading to a real-time edge detector chip. >

A distributed blackboard system for vision applications.

Abstract: This paper describes an implementation of a distributed parallel blackboard system that runs on a Meiko multitransputer system. The blackboard is split up amongst the transputers to allow for distributed local processing, yet the access of the blackboard is transparent to the user processes, irrespective of whether the data is local or remote. Multiple expert processes are invoked as processing resources become available and task dependencies are resolved. The implementation allows both task and data parallelism. A Canny edge detector implementation achieved a speedup of 21 times on 64 transputers.

Multi-Oriented Multi-Resolution Edge Detection

Abstract: In order to build an edge detector that provides information on the degree of importance spatial features represent in the visual field, I used the wavelet transform applied to two-dimensional signals and performed a multi-resolution multi-oriented edge detection. The wavelets are functions well-localized in spatial domain and in frequency domain. Thus the wavelet decomposition of a signal or an image provides outputs in which you can still extract spatial features and not only frequency components. In order to detect edges the wavelet I chose is the first derivative of a smoothing function. I decompose the images as many times as I have directions of detection. I decided to work for the moment on the Xdirection and the Y-direction only. Each step of the decomposition corresponds to a different scale. I use a discrete scale s = 2j (dyadic wavelet) and a finite number of decomposed images. Instead of scaling the filters at each step I sample the image by 2 (gain in processing time). Then, I extract the extrema, track and link them from the coarsest scale to the finest one. I build a symbolic image in which the edge-pixels are not only localized but labelled too, according to the number of appearances in the different scales and according to the contrast range of the edge. Without any arbitrary threshold I can subsequently classify the edges according to their physical properties in the scene and their degree of importance. This process is subsequently intended to be part of more general perceptual learning procedures. The context should be: none or as little as possible a priori knowledge, and the ultimate goal is to integrate this detector in a feedback system dealing with color information, texture and smooth surfaces extraction. Then decisions must be taken on symbolic levels in order to make new interpretation or even new edge detection on ambiguous areas of the visual field. Comments University of Pennsylvania Department of Computer and Information Science Technical Report No. MSCIS-90-13. This technical report is available at ScholarlyCommons: https://repository.upenn.edu/cis_reports/540 Mult i-Oriented Multi-Resolut ion Edge Detection MS-CIS-90-13 GRASP LAB 205

Simultaneous region and edge segmentation of infrared images using non-maximal suppression for edge thinning.

Abstract: A cooccurrence space is defined by utilising the combinations of pixel strengths defined by a Canny edge operator. A region and boundary segmentation derived from this space is first edge thinned by non-maximal suppression and then hysteresis is used as a post-processing step to improve the edges. The distributions in cooccurrence space define the thresholds employed in the hysteresis post-processing.

Use of directional consistency with Sobel edge detector

Abstract: Although there are many edge detection operators that are used in Automatic Target Recognition (ATR) Systems we believe that the required performance can be accomplished using simple operators such as the Sobel operator with minor modification through pre and postprocessing. In this paper we describe two methods that enhance the Sobel performance. The first which can be considered pre processing increases the effective size of the operator to 5*5 The second which is postprocessing modifies the edge magnitude output based on the consistency of the local edge direction. Both methods can be easily implemented on SIMD machines and they are effective in deleting isolated edge points which usually are not part of interesting targets. We will describe the implementation of these techniques on a SIMD machine and study their effect on the performance of an ATR system. 1.