Showing papers in "Pattern Recognition in 1983"

TL;DR: The geodesic distance d x provides rigorous definitions of topological transformations, which can be performed by automatic image analysers with the help of parallel iterative algorithms.

TL;DR: A grayscale digital picture is called connected if it has only one connected component of constant gray level that is maximal, i.e., not adjacent to any component of higher gray level as discussed by the authors.

TL;DR: Results obtained from this study are promising and suggest that closed-loop verification is a potentially more useful technique than previous open-loop processing approaches.

TL;DR: A relaxation method based on patterns of local features is used to find matches between pairs of images or subimages that differ in position or orientation, yielding good results for TV images of objects such as tools and industrial parts, as well as for aerial images of terrain.

TL;DR: A simple, systematic one-pass image segmentation algorithm which is based on the partition mode test of pixels within a (2 × 2) window and assigning and updating label fields to the pixels of this window is described.

TL;DR: A purely data-driven and knowledge-free cluster algorithm is presented which formalizes the Gestalt rule of proximity as it is encountered in dot figures and is discussed in relation to smoothing techniques for the visual enhancement of noisy patterns and to propagation models for the description of visual form.

TL;DR: Approximate locations of axes of symmetry of a 2-dimensional region are detected on the basis of its border because of the way the border is treated in a hierarchical manner where a lower level means a greater number of sides.

TL;DR: A method of classification taking account of a contiguity constraint is presented, applicable to all data sets represented in two distinct spaces, one of which is defined by the introduction of a relation of contiguit.

TL;DR: A k-means clustering algorithm for designing binary tree classifiers is introduced for the classification of cervical cells using the Bhattacharyya distance as a measure of separability between two classes.

TL;DR: Experimental results of cell classification are given to illustrate the efficiency of the proposed method based on multivariate stepwise regression for binary tree classifiers.

TL;DR: F Fourier coefficients of pen-point movement loci relating to strokes are utilized as feature vectors of handwritten characters in an on-line recognition system for handwritten characters based on a Bayesian decision rule.

TL;DR: This paper describes theoretical and experimental results on the use of sequences of interpoint distances extracted from a noisy point image to match the image with one of a set of prototypes.

TL;DR: This paper deals with the decision rules of a tree classifier for performing the classification at each nonterminal node, under the assumption of complete probabilistic information.

TL;DR: The methodology is interesting in itself, as it shows how a geometry can be built-up from a set of labeled counts whose space of definition is shown to be a simplex.

TL;DR: A variant of the original learning rule is analyzed and results are given on its application to the classification of phonemes in automatic speech recognition.

TL;DR: It is shown that binary tree classifiers do not necessarily improve the correct recognition rate in the problem of white blood cell recognition.

TL;DR: Detailed experimental results with remotely sensed multispectral scanner data using the developed model for contextual classification are reported, and classification results do show the effectiveness and efficiency of the developed contextual classifier.

TL;DR: A clustering procedure called HICAP (HIstogram Cluster Analysis Procedure) was developed to perform an unsupervised classification of multidimensional image data and is generalized to process up to 32-bit data with an arbitrary number of dimensions.

TL;DR: The Cox-Lewis statistic leads to one-sided tests for regularity having reasonable power and provides a sharper discrimination between random and clustered data than other statistics.

TL;DR: It is found that methods taking the intensities of the edge pixels into account yield little or no improvement over a simple method of least-squares estimation based only on the locations of these pixels.

TL;DR: A set of “logical pixel logic” based procedures is described and illustrated for detecting the near-minimal number of pixels on a perfectly 8- or 6-connected binary thin line which have to be retained to adequately approximate the original pixel string with a connected sequence of straight-line segments.

TL;DR: An interactive software system that relieves the tedium of visual inspection and standardizes the counting procedure is examined, which includes digitizer, preprocessor and counter subsystems.

TL;DR: A set-theoretical approach in such a way that structural information, defined as classes of patterns having the same structural relationships, induces a redundancy order of codes, so that codes may be reduced automatically from an initial sequential encoding.

TL;DR: MASLOC is a cluster analysis technique based on the p-median model of location theory, which permits for each value of p, clustering the given objects into p groups, each provided with a representative object.

TL;DR: The basic approach is described, using examples from an experimental system being developed for the interpretation of acoustical signals, where knowledge about the physical world that produced the signals is used by the system in the interpretation process.

TL;DR: An algorithm for measuring an index of leakage in fluorescein fundus angiography and an automatic blood vessel elimination technique are presented.

TL;DR: The computational complexity of the hybrid algorithm is shown to be less than that for the Predictor-Corrector Algorithm,(1) for similar performance.

TL;DR: A new clustering algorithm, called CLASSIC, based on an iteratively defined nested sequence of NNRs is presented, which has been applied to various types of gestalt clustering problems and a method for obtaining asymmetric similarities from Euclidean distances is introduced.

TL;DR: The sequentially built recognition scheme has a first stage coarse classification by end points of strokes and a second stage classifier divided into two parts: in one part the decision is based on matching features in the transform space and in the other part a structural classification is applied.

TL;DR: This work uses a texture discrimination system to identify the most likely texture class for a given image given its cluster type and the cluster types of its spatially adjacent neighbors.