Showing papers on "Canny edge detector published in 2000"

On Color Edge Detection.

Abstract: We describe a new color edge detection scheme and a related diffusion process based on hyperbolic coordinates in color space. We derive intensity, hue and saturation based edge detectors and show that the intensity and hue based diffusion processes are governed by the heat equation familiar from conventional scale space theory. A qualitatively new differential equation is derived for the saturation component.

Non-recursive paired tracking for vessel extraction from retinal images

Abstract: This paper reports on a novel vessel tracking algorithm being developped as part of a set of tools for the automated diagnosis of diabetic retinopathy. The algorithm tracks vessels in low-resolution optical images of the retina using their edge map as computed by the Canny edge operator. Tracking proceeds by following an edge line (which coincides with a vein border) while monitoring the connectivity of its twin border. Breaks in the connectivity trigger the creation of seeds that serve as extra starting points for future tracking. Seed creation allows the algorithm to handle bifurcations (a key issue in vessel tracking) and jump over broken or missing edges. Results are presented for ve typical fundus images.

Detecting breast cancer from infrared images by asymmetry analysis

Abstract: Infrared imaging of the breast (also called thermography) has shown effective results in both risk assessment and prognostic determination of breast cancer. This paper proposes an automated approach to detect asymmetric abnormalities in thermograms. Canny edge detector is first used to derive the edges from the original image. Hough transform is then applied to the edge image to recognize the four feature curves, which include the left and the right body boundary curves, and the two parabolic curves indicating the lower boundaries of the breasts. Segmentation is conducted based on the intersection of the two parabolic curves and the body boundaries. Bezier histogram is then derived from each segment. Curvature information is finally computed from the histogram to be used to easily indicate the asymmetry.

A modified fuzzy Sobel edge detector

Abstract: A modified fuzzy Sobel method for edge detection and enhancement is proposed. This method is a modification of the fuzzy Sobel method proposed by Kuo, Lee and Liu see (IEEE Conference on Fuzzy Systems, p.1069-74, 1997). The proposed method overcomes the drawbacks of the conventional gradient methods for edge detection such as Prewitt and Sobel methods. It automatically obtains four threshold values, and apply fuzzy reasoning for edge enhancement. The edges extracted by this method are very clear and provides better representation for image edges and object contours.

Uncertainty characterization in image-based measurements: a preliminary discussion

Abstract: There are a large number of application fields where measurements deriving from digital images can assume a great relevance. Nevertheless, to make a profitable use of such measurements, it is indispensable to achieve a complete and quantitative control on uncertainties that real systems introduce along the chain of steps going from real-world objects to the results of the measurement process. This paper deals with this nontrivial task and, in particular, with the analytical expression of uncertainty characterizing the results of image processing software. At first, a simplified model of uncertainty of digital images is derived and experimentally tested; then the Canny edge detector output uncertainty is analytically expressed and verified both in artificial and real-world images.

Two practical issues in Canny's edge detector implementation

Abstract: We address two practical issues, namely smoothing factor selection and efficient implementation of thresholding with hysteresis, in implementing Canny's edge detector. The smoothing factor of the Gaussian kernel should be chosen to maximize the discrete version of Canny's original criteria. Thresholding with hysteresis should be implemented using an efficient connected component analysis algorithm. Following these suggestions in implementing Canny's edge detector will in general result in optimal edge detection quality and very significant reduction in running time for large images.

On the bias of the likelihood ratio edge detector for SAR images

Abstract: The likelihood ratio edge detector is an efficient filter for the segmentation of SAR images The authors show that this filter provides biased location of the edge if it is not an "ideal" step edge of known orientation A simple model enables the authors to interpret this observation and to evaluate the bias

A normalized color difference edge detector based on quaternion representation

Abstract: This paper presents the quaternion color difference edge detector, a new approach to detection of edges in color images. Based on a new type of convolution, the color difference subspace and the proposed edge detector are expressed analytically by using the algebra of the quaternion. The proposed color image edge detector generates edges only where sharp changes of color occur in the original image. The experimental results have verified the improved performance of the new edge detector compared to some well known methods.

On learning texture edge detectors

Abstract: Texture is an inherently non-local image property All common texture descriptors, therefore, have a significant spatial support which renders classical edge detection schemes inadequate for the detection of texture boundaries In this paper we propose a novel scheme to learn filters for texture edge detection Textures are defined by the statistical distribution of Gabor filter responses Optimality criteria for detection reliability and localization accuracy are suggested in the spirit of Canny's edge detector Texture edges are determined as zero crossings of the difference of the two a posteriori class distributions An optimization algorithm is designed to determine the best filter kernel according to the underlying quality measure The effectiveness of the approach is demonstrated on texture mondrians composed from the Brodatz album and a series of synthetic aperture radar (SAR) imagery Moreover, we indicate how the proposed scheme can be combined with snake-type algorithms for prior-knowledge driven boundary refinement and interactive annotation

Inconsistencies in edge detector evaluation

Abstract: In recent years, increasing effort has gone into evaluating computer vision algorithms in general, and edge detection algorithms in particular. Most of the evaluation techniques use only a few test images, leaving open the question of how broadly their results can be interpreted. Our research tests the consistency of the receiver operating characteristic (ROC) curve, and demonstrates why consistent edge detector evaluation is difficult to achieve. We show how easily the framework can be manipulated to rank any of three modern edge detectors in any order by making minor changes to the test imagery. We also note that at least some of the inconsistency is the result of the erratic nature of the algorithms themselves, suggesting that it is still possible to create better edge detectors.

Edge Thinning Based on Sobel Operator

Abstract: A new edge thinning method based on Sobel operator is proposed in this paper. A gray edge image (P1) is obtained by introduces a attenuation factor,the gray edge image can also be processed using Sobel operator,then another gray edge image (P2) is get from the previous gray edge image. Another gray edge image(P3) is obtained by subtracting P2 from P1 and the pixels of negative value in P3 are set to zero to eliminate the extra points on both side of the detected edges therefor a image with thinner edge can be obtained. And this procedure can repeat several times to obtain thinner edges(but not necessarily continuous). This method also has effect on the edges that detected by other edge detection methods. Edge thinning can be used to get the exact position of object's edge. This method can not only get fairly thin edge but can also reserve the edges of low step change and this advantage can overcome the segmentation error that due to improperly selected threshold.

Edge detection by genetic algorithm

Abstract: This paper presents a novel edge detection method based on a genetic algorithm (GA). Only a simple image data set is required for constructing the edge filter. The data set, which consists of a simple image and its expected edge features, is used for training by the GA. The merit of this method is that an effective edge extraction filter can be easily constructed. Results and examples are illustrated in the paper.

Grouping edge points into line segments by sequential Hough transformation

Abstract: An algorithm to group edge points into digital line segments with Hough transformation is described. The edge points are mapped onto the parameter domain discretized at specific intervals, on which peaks appear to represent different line segments. By modeling each peak as a Gaussian function in the parameter domain, a region to which the edge points are supposed to be mapped is determined. Then the edge points are grouped and the parameters for a line segment are computed. For the edges including multiple line segments, a sequential Hough transformation for detecting peaks one by one in the parameter domain is implemented, and the points from the region around each peak are grouped, thus the line segments are described. Experiments show the robustness of the algorithm implemented on both the generated edges disturbed by different noise levels and real images taken from an indoor environment.

Edge detection from noisy images using a neural edge detector

Abstract: In this paper, a new edge detector using a multilayer neural network, called a neural edge detector (NED), is proposed for detecting the desired edges clearly from noisy images. The NED is a supervised edge detector: through training the NED with a set of input images and desired edges, it acquires the function of a desired edge detector. The experiments on the NED to detect the edges from noisy test images and noisy natural images were performed. By comparative evaluation with the conventional edge detectors, the following has been demonstrated: the NED is robust against noise; the NED can detect clear continuous edges from the noisy images; and the performance of the NED is the highest in terms of similarity to the desired edges.

Page segmentation and text extraction from gray-scale images in microfilm format

Abstract: The paper deals with a suitably designed system that is being used to separate textual regions from graphics regions and locate textual data from textured background. We presented a method based on edge detection to automatically locate text in some noise infected grayscale newspaper images with microfilm format. The algorithm first finds the appropriate edges of textual region using Canny edge detector, and then by edge merging it makes use of edge features to do block segmentation and classification, afterwards feature aided connected component analysis was used to group homogeneous textual regions together within the scope of its bounding box. We can obtain an efficient block segmentation with reduced memory size by introducing the TLC. The proposed method has been used to locate text in a group of newspaper images with multiple page layout. Initial results are encouraging, we would expand the experiment data to over 300 microfilm images with different layout structures, promising result is anticipated with corresponding modification on the prototype of former algorithm to make it more robust and suitable to different cases.

Improving image segmentation using edge information

Abstract: We report methods for image segmentation that combine region growing and edge detection. Existing schemes that use region-based processing provide unambiguous segmentation, but they often divide regions that are not clearly separated, while merging regions across a break in an otherwise strong edge. Edge-based schemes are subject to noise and global variation in the picture (e.g. illumination), but do reliably identify strong boundaries. Our combined algorithm begins by using region growing to produce an over-segmented image. This phase is fast (order N, where N is the number of pels in the image). We then modify the over-segmented output of the region growing using edge criteria such as edge strength, edge smoothness, edge straightness and edge continuity. Two techniques-line-segment subtraction and line-segment addition-have been investigated. In the subtraction technique, the weakest edge (based on a weighted combination of the criteria) is removed at each step. In addition technique, the strongest edge is used to seed a multi-segment line that grows out from it at both ends. At every junction, the adjoining edge that has the highest edge strength is appended. We have also investigated a form of look-ahead, where the growing of lines depends on the strength of the adjoining edge and those to which it is linked. The overall procedure for both techniques, current results and the areas for improvement and expansion have been discussed.

Optimal line detector

Abstract: An optimal line detector for the one-dimensional case is derived from Canny's criteria (1986). The detector is extended to the two-dimensional case by operating separately in the x and y directions. An efficient implementation using an infinite impulse response (IIR) filter is provided. This implementation has an additional advantage that increasing the filter scale affects neither temporal nor spatial complexity. Our detector is faster than the Gaussian used by Steger (1998); e.g., when the scale is 3 our detector is 33 times faster. Experimental results using real images demonstrate the validity of the algorithm.

Engineering design of an image acquisition and analysis system for dendrochronology

Abstract: The design and implementation of a computer-vision-based analysis system for dendrochronology are described. The primary issues covered are automatic focus detection, illumination distortion, spatial and radiometric registration of a sequence of images into a single mosaic, and edge detection and linking. These issues are not unique to the ap- plication, but are likely of interest to anyone developing automated image analysis systems. A modification of the Canny edge detection algorithm that adapts to the variable characteristics of tree-ring images is also described. © 2000 Society of Photo-Optical Instrumentation Engineers. (S0091-3286(00)01502-6)

Refining edge detection within spiral architecture

Abstract: Detection of edge points of 3-dimensional physical object in a 2-dimensional image is one of the main research areas of computer vision. Object contour detection and object recognition rely heavily on edge detection. The Spiral Architecture, a relatively new image structure, possesses powerful computational features that are pertinent to the vision process. In this paper, we present an edge detection scheme using Gaussian Multi-resolution Theory. The detection algorithm reduces noise and unnecessary details of the image. This algorithm is applied to obtain the edge maps of some well-known images for image processing.

Efficacy of Gaussian smoothing in Canny edge detector

Abstract: The authors examine the efficacy of the noise-filtering stage in the Canny edge detector using the receiver operating characteristic paradigm and find that omitting the filtering stage can give superior results. They conclude that the non-maximal suppression stage is mainly responsible for the success of the Canny detector.

Human visual system singularity map analyses

Abstract: This paper has analyzed the day and night images using the novel concept called hard and soft singularity maps (SM) that are biologically extracted by the lateral redundant data. Consequently, the correspondence exists uniquely among neighborhood frames in terms of the different slope values at image corners solving the optical flow problem for the video compression. In this paper, some efficient computational methods: min-max picking, and next order Cellular Neural Network implementing the anti-diffusion Laplacian, can obtain the SM without the convolution broadening based on Sobel and Canny edge operators. However, the differentiation operation may produce a false singularity under noise, and thus we apply the Hermitian wavelets to obtain the noisy singularity.

Image edges from area morphology

Abstract: This paper introduces an edge detection process based on area morphology Area open-close and area close-open operators are used to generate scaled image representations for feature extraction The edges are defined by the boundaries of the scaled objects in the area-filtered images From the area open-close and close-open operators, thin closed contours suitable for image segmentation are produced The edge maps allow exact specification of the minimum area for the extracted regions and are Euclidean invariant and causal through scale space Results are given that demonstrate the effectiveness of the area operator-based edge detection In contrast to traditional edge detectors, edge detection via area morphology provides well-localized boundaries and does not require thresholding

The advances of edge-detection in medical images

Abstract: Edge -detection operators and multi-SE edge pyramid are in troduced systemly, and the experimental results of several good operators is given.

Edge Detection of SAR Images Based on the Fractal Theory

Abstract: After discussing the SAR images' properties and the principles of image description with discrete Fractional Brownian incremental random field model(DFBR model),this paper presents two new methods of edge detection for SAR images based on the Fractal theory It is known that the Fractal Brownian random model is valid to describe the image of nature sceneIn one tiny region of image, the gray surface is self similar to statisticsBut for the edge points ,the point located in the boundaries between adjacent regions,this property will be lostAnd the Fractal parameters of these points will be out of rangeSo we can determine whether the point of image is edge point via calculating its Fractal parametersA new fast method to calculate the Fractal dimension and the principles of determining edge point based on single Fractal dimension are discussed in this paper Moreover,the Fractal parameters of image points must not depend on the scaleBut for the edge points,because they have lost the self similar,their Fractal parameters will change more quickly with the scales than the other pointsSo we can estimate the points' Fractal parameters in different scales and caculate their difference matrix(edge magnitude map)Then we can obtain binary edge image by thresholding this edge map In the end,the authors present experimental results of edge detection of a SAR images with the Sobel operator,the Prewitt operator,and these two methods are mentioned aboveAfter analyzing the experimental results in detail,the authors draw the conclusion: these two new methods have better ability to anti noise for SAR images edge detectionAnd the second method,the Multifractal Method,get better results of edge detection than the first method,the single fractal method

Optimal shape detection

Abstract: We present a new approach for accurate detection of two-dimensional shapes. We first derive an optimal smoothing filter, which minimizes both the noise power and the mean squared error between the input and the filter output. This operator is found to be a derivative of the double exponential (DODE) function. We define an operator for shape detection by extending the DODE filter along the shape's boundary contour. We find that this filtering scheme is equivalent to integrating gradients along the hypothetical shape boundary, but our method turns out to be more robust than conventional edge detection followed by edge grouping. This approach also provides a tool for a systematic analysis of edge-based shape detection. We investigate how the error is propagated by the shape geometry. This enables us to predict both its localization and detection performance. Application to vehicle detection in aerial images and human facial feature detection are provided.

Development of the Technology of Edge Detection

Abstract: Edge detection is an indispensable step before image analysis and recognition, and is an important technology in the image preprocessing procedure. In this paper we chiefly review the study of the edge detection methods , and then disguss the characteristic of some typical edge detection algorithms, analyze their qualities, and finally give the development of edge detection methods in the future.

Automatic facial feature detection for model-based coding

Abstract: This paper presents an automatic facial feature detection system for 3D model based coding applications. This proposed system is based on simple image processing techniques, which can be easily implemented using parallel algorithms is parallel processing hardware. Model Based Face Coding can be used in remote teaching to enhance the quality of remote teaching, where by reducing the barrier between teacher and student. In this case, only a selected set of control points of the face is transmitted to the remote terminal instead of sending video signal. In order to extract this set of control points a predefined 3D generic wire frame model is used. In this paper, automatic extraction process of the feature points of facial images needed for 3D model fitting is discussed. The proposed detection methods for all the facial features utilize filtering, thresholding, edge detection and edge counting without any manual adjustments or initialization. Head top, chin points, eye center, mouth center and nose center were detected using vertical integral projection method. The centroid method was used successfully for eyebrow center detection. Four points of the mouth features were detected with both Canny edge detection method and amplitude projection method. The first one had limited success and second gave very satisfactory results. On the whole, the results obtained are encouraging and could be used in automatic registration of 2D facial images into 3D face models. Subsequent tracking of some of these feature points lead us to automatic facial expression recognition using optical flow.