Showing papers on "Corner detection published in 1998"

Feature Detection with Automatic Scale Selection

Abstract: The fact that objects in the world appear in different ways depending on the scale of observation has important implications if one aims at describing them. It shows that the notion of scale is of utmost importance when processing unknown measurement data by automatic methods. In their seminal works, Witkin (1983) and Koenderink (1984) proposed to approach this problem by representing image structures at different scales in a so-called scale-space representation. Traditional scale-space theory building on this work, however, does not address the problem of how to select local appropriate scales for further analysis. This article proposes a systematic methodology for dealing with this problem. A framework is presented for generating hypotheses about interesting scale levels in image data, based on a general principle stating that local extrema over scales of different combinations of γ-normalized derivatives are likely candidates to correspond to interesting structures. Specifically, it is shown how this idea can be used as a major mechanism in algorithms for automatic scale selection, which adapt the local scales of processing to the local image structure. Support for the proposed approach is given in terms of a general theoretical investigation of the behaviour of the scale selection method under rescalings of the input pattern and by integration with different types of early visual modules, including experiments on real-world and synthetic data. Support is also given by a detailed analysis of how different types of feature detectors perform when integrated with a scale selection mechanism and then applied to characteristic model patterns. Specifically, it is described in detail how the proposed methodology applies to the problems of blob detection, junction detection, edge detection, ridge detection and local frequency estimation. In many computer vision applications, the poor performance of the low-level vision modules constitutes a major bottleneck. It is argued that the inclusion of mechanisms for automatic scale selection is essential if we are to construct vision systems to automatically analyse complex unknown environments.

Robust image corner detection through curvature scale space

Abstract: This paper describes a novel method for image corner detection based on the curvature scale-space (CSS) representation. The first step is to extract edges from the original image using a Canny detector (1986). The corner points of an image are defined as points where image edges have their maxima of absolute curvature. The corner points are detected at a high scale of the CSS and tracked through multiple lower scales to improve localization. This method is very robust to noise, and we believe that it performs better than the existing corner detectors An improvement to Canny edge detector's response to 45/spl deg/ and 135/spl deg/ edges is also proposed. Furthermore, the CSS detector can provide additional point features (curvature zero-crossings of image edge contours) in addition to the traditional corners.

Gray Level Corner Detection.

Abstract: In this paper the analysis of gray level corner detections has been carried out. The performances of cornerness measures for some corner detection algorithms are discussed. This paper presents a new approach for corner detection called the GradientDirection corner detector which is developed from the popular Plessey corner detector. The GradientDirection corner detector is based on the measure of the gradient module of the image gradient direction and the constraints of false corner response suppression.

Corner detection device and corner detection method

Abstract: A corner detector capable of detecting each corner position of a predetermined area from an input signal with a simple structure. There is provided memory means (9B) for storing memorizing an input signal (keyT) and corner detection means (9C) for detecting a point where the signal level becomes the reference signal level or more at first by reading an input signal stored in the memory means sequentially in horizontal direction from an upper limit and a lower limit of a retrieval scope and in vertical direction from the left end to the right end of the retrieval scope, and detecting a point where the signal level becomes the reference signal level or more at first by reading an input signal stored in the memory means sequentially in a diagonal direction at a predetermined angle from each angle of the retrieval scope to detect from the detected points four points whose positions differ. With such means, each corner position of the area of the detection object can be detected with a simple structure and quickly.

Morphological corner detection

Abstract: This paper presents a new operator for corner detection. This operator uses a variant of the morphological closing operator, which we have called asymmetrical closing. It consists of the successive application of different morphological transformations using different structuring elements. Each of these structuring elements used to probe the image under study is tuned to affect corners of different orientation and brightness. We found that this kind of approach, based on brightness comparisons, leads to better quality results than others and is achieved at a lower computational cost.

Moment and curvature preserving technique for accurate ellipse boundary detection

Abstract: Circles and their elliptic projections are very commonly used image features in computer vision applications. Thus, it is very important to be able to determine their location in an accurate manner. A technique for determining an ellipse boundary with subpixel precision is proposed. The technique, called the moment and curvature preserving detection (MCP), utilizes the first three intensity moments and the intensity gradient of the image. The idea of using moments for subpixel edge detection is not new, but in the case of ellipses the moments do not provide sufficient information for precise detection. However, if the local curvature is augmented to the observations, the ellipse boundary can be determined reliably.

Curvature scale space for robust image corner detection

Abstract: This paper describes a new method for image corner detection based on the curvature scale space (CSS) representation. The first step is to extract edges from the original image using a Canny detector. The corner points of an image are defined as points where image edges have their maxima of absolute curvature. The corner points are detected at a high scale of the CSS image and the locations are tracked through multiple lower scales to improve localization. The CSS corner detector is very robust to noise and performed better than three other detectors it was compared to.

A mean field annealing approach to robust corner detection

Abstract: This paper is an extension of our previous paper to improve the capability of detecting corners. We proposed a method of boundary smoothing for curvature estimation using a constrained regularization technique in the previous paper. We propose another approach to boundary smoothing for curvature estimation in this paper to improve the capability of detecting corners. The method is based on a minimization strategy known as mean field annealing which is a deterministic approximation to simulated annealing. It removes the noise while preserving corners very well. Thus, we can detect corners easier and better in this approach than in the constrained regularization approach. Finally, some matching results based on the corners detected by corner sharpness in the mean field annealing approach are presented as a demonstration of the power of the proposed algorithm.

Corner detection using Gabor-type filtering

Abstract: In this paper a new corner detection technique using Gabor filters is proposed. It is suitable for both binary and gray level images with varying backgrounds. This technique is based on a scale interaction model, where the difference of two lowpass filters (with different bandwidths) is utilized to extract the corners, line intersections and line endings in the input image. The filtering is done iteratively until the change in the output is below a certain threshold. This technique does not require tracking the boundary and computing the curvature. The important features of the scheme include simplicity, robustness, speed of processing and availability of simple controls to tune the technique for various computer vision applications.

Curvature scale space based image corner detection

Abstract: This paper describes a new method for image corner detection based on the curvature scale space (CSS) representation. The first step is to extract edges from the original image using a Canny detector. The Canny detector sometimes leaves a gap in T-junctions so during edge extraction, the gaps are examined to locate the T-junction corner points. The corner points of an image are defined as points where image edges have their maxima of absolute curvature. The corner points are detected at a high scale of the CSS and the locations are tracked through multiple lower scales to improve localization. The final stage is to compare T-junction corners to CSS corners and remove duplicates. This method is very robust to noise and we believe that it performs better than the existing corner detectors.

Corner Detection Via Topographic Analysis of Vector Potential.

Abstract: This paper describes how corner detection can be realised using a new feature representation based on a magneto-static analogy. The idea is to compute a vector-potential by appealing to an analogy in which the Canny edge-map is regarded as an elementary current density residing on the image plane. In this paper, we demonstrate that corners are located at the saddle-points of the magnitude of the vector-potential. These points correspond to the intersections of saddle-ridge and saddle-valley structures, i.e. to junctions of the edge and symmetry lines. We describe a template-based method for locating the saddle-points. This involves performing a non-minimum suppression test in the direction of the vector-potential and a non-maximum suppression test in the orthogonal direction. Experimental results using both synthetic and real images are given. We investigate the angle and scale sensitivity of the new corner detector and compare it with a number of alternative corner detectors.

6-2 Gray Level Corner Detection

Abstract: In this paper the analysis of gray level corner detections has been carried out. The performances of cornerness measures for some corner detection algorithms are discussed. This paper presents a new approach for corner detection called the GradientDirection corner detector which is developed from the popular Plessey corner detector. The GradientDirection corner detector is based on the measure of the gradient module of the image gradient direction and the constraints of false corner response suppression.

Image processing system for direction detection of an object using neural network

Abstract: The purpose of this study is to develop a high-speed and robust image processing system composed of neural network for detecting and discriminating twenty directions of an object which is randomly laid and piled up on a horizontal plane. Our proposed direction detection system was designed to consist of two modules of neural networks; one is an edge detection module and the other is a direction detection module. An edge image is extracted from an original image taken with a CCD camera and the object direction is detected from the edge image. The edge detection module could detect edges when differences of gray level between the object and the background were larger than thirteen at any gray level of the object. Discrimination ratio of object direction was successful at 93.5%, when one object was laid on an arbitrary place in the area.

Corner detection using vector potential

Abstract: This paper describes how corner detection can be realised using a new feature representation that has recently been successfully exploited for edge and symmetry detection. The feature representation based on an magneto-static analogy. The idea is to compute a vector potential by appealing to an analogy in which the Canny edge-map is regarded as an elementary current density residing on the image plane. In our previous work we demonstrated that edges are the local maxima of the vector potential while points of symmetry correspond to the local minimum. In this paper we demonstrate that corners are located at the saddle points of the magnitude of the vector potential. These points corresponds to the intersections of saddle-ridge and saddle-valley structures, i.e. to junctions of the edge and symmetry lines. We describe a template-based method for locating the saddle-points. This involves performing a nonminimum suppression test in the direction of the vector potential and a nonmaximum suppression test in the orthogonal direction. Experimental results of both synthetic and real images are given.

Handwritten Chinese signature verification based on attributed string matching of stroke linkage order

Abstract: We propose an attributed string matching approach based on the writing sequences of an input signature for Chinese signature verification. It is impossible to find features of a signature that are invariant with respect to individual writing style. The aim of our research is to find a particular feature set that will exhibit small intraclass variance. A signer tends to connect consecutive strokes in a constant sequences when signing a signature, especially for Chinese signatures. Therefore, the writing sequences (stroke order) of a signature can be regarded as the personal signature feature. In order to obtain an attributed string that will be used in the string matching similarity calculation, we must split an input signature into several segments from the corners of strokes at first. Since the wavelet transform has been used in the field of edge detection and corner detection for a long time, we can use it as a tool for corner detection to find an optimal segmentation set. Therefore, an attributed string for consecutive segments can be calculated after the signature has been split into several segments suitably by means of a wavelet transform. The elements in the attributed vector include relative angle of adjoining segments, the direction code of the segment, writing duration time of the segment, and the length of the segment. The total relative angle can also be summarized over all of the relative angles of segments to result in another individual feature. This total relative angle can be used to filter the rougher forgery signatures. Our attributed string matching method is also based on the extraction of irreducible characteristic points. The experimental results show a very excellent discrimination capability.

Directional second order derivatives: Application to edge and corner detection

Abstract: In this paper, we propose an adaptive scheme to design directional second order derivatives orthogonally and tangentially to the local edge. The principle lies on the definition of two adaptive filter masks which estimate the two derivatives along the normal (n) and tangential (t) directions. Both filter masks are controlled by an adaptive mask whose coefficients are tuned in accordance with the local grey level distribution. The two new filters are then applied respectively to edge and corner detection : edge detection is achieved by detecting the zero-crossing of the derivative along n. and corner detection is obtained by thresholding the amplitude of the derivative along t. Results of these detections are provided on synthetise and real-world images, and swow the robustness of the new proposed approach.

Applications of the "Creep-and-Merge" System: Corner Detection

Abstract: The Creep‐and‐Merge (CAM) segmentation system (described in [3, 2]) is a novel architecture for region-based segmentation; it is designed to be efficient, insensitive to noise, scale and image geometry, capable of applying the widest range of statistical models, and to contain no adjustable parameters. This paper describes the continuing development of higher‐level processing models utilising this system: we present a geometric (“true”) corner detector with good performance, and give qualitative and quantitative comparisons with other leading systems.

Contour tracer for a fast and precise edge-line extraction

Abstract: Feature extraction algorithms delivering accurate results are computational intensive and cannot keep up with the demanded processing times on single processor systems in automated micropart assembly tasks. For this case an improved contour tracer for real-time edge-line extraction, using adaptive local thresholds, is introduced. In combination with an effective start point search, that concentrates on the strongest local contours, and a two stage robust corner detection a fast and precise edge-line extraction comes possible.

Quadric-matching-based reconstruction of space curve

Abstract: A fast 3D reconstruction of space curve based on quadric segment and matching in trinocular vision is proposed in this paper. First by using the perspective invariance of zero curvature and corner points in curves, curves can be segmented into different quadric curves. Then match these segments among three images and reconstruct them respectively. Finally combine all reconstructed segments together to form continuous curves. The experiments illustrated that the reconstruction speed is higher than those which take points or lines as primitives and duality and a uniqueness solution can be ensured.© (1998) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Weighted statistical model for non-rigid image objects detection

Abstract: A new weighted statistical model is presented for detecting non-rigid image targets In the detection or matching stage, a modified GA (genetic algorithm) is presented to solve the problem of adaptability to scaling, position and tilted angles We demonstrate our WSM model on the practical human face detection problem Some results are presented

Inspection of power and ground layers in PCB images

Abstract: In this work, we present an inspection method for power and ground (PG (2) decomposition of contours into segments: corner detection and matching lines or circular arcs between two corners; (3) determination of the pads from partial contour information obtained in step (2), and (4) design rules checking for each detected pad.© (1998) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

A Generalisation Study on Neural Corner Detection

Abstract: Dominant Point Detection (DPD) is one of the tasks in image analysis; it aims making polygonal approximations through the search of a set of points of relevance in a contour, reducing the amount of information. In this work, the ability of neural networks to learn the performance of several DPD algorithms is studied. For it a dynamic neural net that traverses the contour will be used, giving a relevance measurement for each point and detecting them through a simple post-processing phase. Different training sets and net configurations were used. The results of applying the neural algorithm to images of real objects show its validity, and also the ability of neural nets to learn previously unknown DPD algorithms.