Showing papers in "Pattern Recognition Letters in 2000"

TL;DR: A Catmull–Rom spline-based lane model which describes the perspective effect of parallel lines has been proposed for generic lane boundary and its coarse-to-fine matching offers an acceptable solution at an affordable computational cost, and thus speeds up the process of lane detection.

TL;DR: The combination of neural and statistical algorithms is proposed as a method to obtain high accuracy values after much shorter design phases and to improve the accuracy–rejection tradeoff over those allowed by single algorithms.

TL;DR: A new method based on shape information that can be very efficient for images with simple background, based on edge detection based on a multiple-scale filter for automatic detection of human faces.

TL;DR: The results clearly indicate that tool condition monitoring which is defined as the ability to distinguish between a sharp, a semi-dull, or a dull tool can be successfully accomplished by combining sensory data from a CCD camera and a microphone.

TL;DR: This paper presents a novel pattern recognition application, the application in agronomy and proposes a roundness measurement concept to analyze the roundness of rose flower shapes for the new application and shows that this measurement is an efficient parameter for image shape analysis.

TL;DR: A new dichotomization technique is proposed for multilevel thresholding based on selection of the consistent peak location of the correlation function as threshold value over the interested histogram region that gives consistent results in the sense of human perception and gives satisfactory results to find uniform regions in the image plane.

TL;DR: A simple yet effective method for color texture classification that uses the CIE xy chromaticity diagram of an image and a corresponding set of two-dimensional and three-dimensional moments to characterize a given color texture.

TL;DR: A class-dependent weighted dissimilarity measure in vector spaces is proposed to improve the performance of the nearest neighbor (NN) classifier and an approach based on Fractional Programming is presented.

TL;DR: A new deterministic crossover method based on the pairwise nearest neighbor method is introduced that shows that high quality codebooks can be obtained within a few minutes instead of several hours as required by the previous GA-based methods.

TL;DR: This method is extended to the case where dissimilarities are only known to lie within certain intervals, and shows the ability of this method to represent both the structure and the precision of dissimilarity measurements.

TL;DR: Experimental results reveal the feasibility and superiority of the proposed approach in solving color quantization problem and the executing speed of the algorithm is quite fast due to the reduced RGB color space, sorted histogram list, suitable color design and destined pixel mapping.

TL;DR: It will be shown that under a particular set of cost functions for any two graphs, g and g', a maximum common subgraph is a mean and any subgraph of g or g' that contains amaximum common sub graph is a means as well.

TL;DR: Two new evaluation methods for edge detectors are presented, non-contextual and concerns the evaluation of edge detector performance in terms of detection errors and contextual method, which evaluates the performance of edge detectors in the context of image reconstruction.

TL;DR: This work uses blur invariant moments to normalize the image of the object to a canonical form with respect to an assumed affine motion and obtains the motion parameters from the normalizing transformation matrices using moment functions.

TL;DR: This work proposes to use an approximate Hausdorff measure in the dimension D to estimate the complexity of non-fractal objects to reduce the uncertainty in the divider-step method.

TL;DR: A new method to divide a planar curve into circular arcs in order to recognize the overlapped particles in an image is described, based on a new breakpoint detection method.

TL;DR: This paper proposes to perform the optimization of the BN parameters using an alternative approach to the EM algorithm: the BC + EM method, and provides experimental results to show that this proposal results in a more effective and efficient version of the Bayesian Structural EM algorithm for learning BNs for clustering.

TL;DR: An algorithm that operates on a bit-mapped text pattern array to determine the up/down orientation of the page, that is, whether the page is upright or inverted, is presented.

TL;DR: A genetic algorithm (GA) is proposed which will allow the system to combine different solutions in a stochastic search so as to produce better results and has proven to converge faster than the first one, although the final results reached are very similar to each other.

TL;DR: From the experimental results, it has been confirmed that the proposed system enables high quality grading of beef marbling, and robust region segmentation of the actual beef rib-eye image into lean and fat regions.

TL;DR: This paper describes an efficient approach for the detection of line junctions and line terminations using a local measure of line curvature and validated on several synthetic and real images.

TL;DR: An algorithm for the reconstruction of 3D medical objects from ultrasound images using a Rayleigh model to describe the image formation process and an optimal Bayesian criterion is described.

TL;DR: A simple and yet robust Hough transform algorithm is proposed to detect and analyze reflectional symmetry and skew-symmetry (reflectional symmetry under parallel projection) under the presence of noise and occlusion.

TL;DR: This work attempts to distinguish land from water in satellite images, specifically images taken by the FORTE satellite, and successfully approximate areas hidden by stationary artefacts in the image.

TL;DR: A multiscale roughness descriptor for particles is proposed, based on the harmonic wavelet transform, that acts as a mathematical microscope and allows analyses of wear and erosion phenomena acting on particles.

TL;DR: The experimental results showed that this refined technique was successful in classifying the objects under consideration and that it outperformed those traditional methods especially in the presence of noise.

TL;DR: In this article, the structural characteristics between background regions and character components are investigated as a novel approach to cursive script segmentation, which achieved around 80% correct segmentation on a difficult database of unconstrained handwritten words.

TL;DR: This paper shows how to detect families of parallel lines in 3-D space at a moderate computational cost by using a (2+2)-D Hough space by finding peaks in the 2-D slope parameter space.

TL;DR: It has been found that the bestOctagonal distance in 2D performs equally good with respect to the best one for the weighted distances (〈3,4〉) and in 3D, the octagonal distance {1,1,3} has an edge over the other good weighted distances.

TL;DR: It is proposed in this paper that based on the four fuzzy measures, namely, linear and quadratic index of fuzziness, logarithmic and exponential entropy measures, how the best threshold will be identified and used.