scispace - formally typeset
Search or ask a question

Showing papers in "Pattern Recognition in 1980"


Journal ArticleDOI
TL;DR: The ability of the RNG to extract a perceptually meaningful structure from the set of points is briefly discussed and compared to that of two other graph structures: the minimal spanning tree and the Delaunay (Voronoi) triangulation.

1,213 citations


Journal ArticleDOI
TL;DR: A new technique for fast “scan-along” computation of piecewise linear approximations of digital curves in 2-space is described and the application to the boundaries of the images of a lung and a rib in chest radiographs is illustrated.

359 citations


Journal ArticleDOI
TL;DR: The functions of minπ and maxπ are introduced as the analogues of nearest neighbour “propagation” signals of binary images as well as extending some already well known binary processes into grey level algorithms.

162 citations


Journal ArticleDOI
TL;DR: An overview of the conception, development, evaluation and documentation of this spectral-spatial classifier for earth observational image data shows it to have improved accuracy, with greater computation efficiency, and only slightly increased operator complexity.

147 citations


Journal ArticleDOI
John Cerella1
TL;DR: Pigeons were trained to identify Charlie Brown in “Peanuts” cartoons and were subsequently treated by the pigeons as equivalent to intact versions, suggesting that pigeons respond to complex line projections as collections of local features rather than as representations of three dimensional objects.

130 citations


Journal ArticleDOI
TL;DR: An efficient procedure which integrates feature selection and binary decision tree construction is presented which yields an optimal classification decision at each node based on the Kolmogorov-Smirnov criterion.

109 citations


Journal ArticleDOI
TL;DR: Monte Carlo methods were used to estimate the percent misclassification of 13 clustering methods for six types of parameterizations of two bivariate normal populations and the overall poorest methods were judged to be nearest neighbor and maximum likelihood.

106 citations


Journal ArticleDOI
TL;DR: The results indicated that theRank order performance of the four methods differed markedly from the rank order generally found in multivariate normal mixture model studies.

101 citations


Journal ArticleDOI
TL;DR: A hierarchically structured recognition system consists of a conventional statistical classifier, a structural classifier analysing the topological composition of the patterns, a stage reducing the number of hypotheses made by the first two stages, and a mixed stage based on a search for maximum similarity between syntactically generated prototypes and patterns.

89 citations


Journal ArticleDOI
TL;DR: The shape number is a description that changes with skewing, anisotropic dilation and mirror images, as the intuitive psychological concept of “shape” demands, Nevertheless, at the end of the paper a related Theory “B” of shapes is introduced that allows anisotrop changes of scale, thus permitting for instance a rectangle and a square to have the same B shape.

87 citations


Journal ArticleDOI
TL;DR: Monte Carlo results presented here further confirm the relatively good performance of non-parametric Bayes theorem type algorithms compared to parametric (linear and quadratic) algorithms and point out certain procedures which should be used in the selection of the density estimation windows for non- Parametric algorithms to improve their performance.

Journal ArticleDOI
TL;DR: The algorithm uses topological matching procedure to compute the correlation coefficients and then maximises the correlation coefficient and some pre-processing techniques are discussed to convert the input image to a labelled graph.

Journal ArticleDOI
TL;DR: A simple convex hull method used for extracting concavity and convexity is described, which was applied to data consisting of 16,400 characters and achieved a very high level of performance.

Journal ArticleDOI
Youji Fukada1
TL;DR: Two clustering procedures for region analysis of image data are described and the security of these algorithms theoretically and examples are presented in order to show how these algorithms work for real image data.

Journal ArticleDOI
TL;DR: The stability of a hierarchical structure with a variation on a measure of stability proposed in the literature is studied, appropriate for proximity matrices whose entries are on an ordinal scale.

Journal ArticleDOI
TL;DR: A computationally efficient agglomerative clustering algorithm based on multilevel theory is presented and it is observed that the savings, in storage space and computation time, increase nonlinearly with the sample size.

Journal ArticleDOI
TL;DR: In segmenting an image by pixel classification, the sequence of gray levels of the pixel's neighbors can be used as a feature vector that yields classifications at least as good as those obtained using other local properties as features.

Journal ArticleDOI
TL;DR: It is emphasized that the interactive pattern recognition system ISPAHAN is well suited to find optimal decision functions based on mappings, and two families of new mapping algorithms are defined.

Journal ArticleDOI
TL;DR: In this paper, random walks are simulated for plane domains A bounded by absorbing boundaries Γ, and the absorption distributions are estimated. Measurements derived from the above distributions are the features used for texture classification.

Journal ArticleDOI
TL;DR: To gain an understanding of the learning process, the author derives expressions for success probability as a function of training time for a one-dimensional increment error correction classifier with imperfect labels.

Journal ArticleDOI
TL;DR: This work defines digital convexity for 4-connected sets of points on a square lattice and shows that a cellular complex is cellularly convex if and only if the digital region determined by the complex is digitally convex.

Journal ArticleDOI
TL;DR: It is shown here that the Hough transform may be used for encoding of line curves and waveforms that consist of the concatenation of curves from an underlying set of families of curves.

Journal ArticleDOI
TL;DR: A definition of a global criterion on a set of objects to be labeled that combines both ambiguity and consistency with adjustable weights is proposed and a projected gradient algorithm is developed to minimize the criterion.

Journal ArticleDOI
TL;DR: Results for more highly structured problems, involving specific covariance matrices, show that in some cases increasing correlation between the measurements yields higher values of P cr, androximate expressions are derived relating P cr dimensionality, training sample size and the structure of the underlying probability density.

Journal ArticleDOI
TL;DR: An adaptive model for computer recognition of vowel sounds with the first three formants as features using a single pattern training procedure for self-supervised learning and maximum value of fuzzy membership function is the basis of recognition.

Journal ArticleDOI
TL;DR: Discrete Karhunen-Loeve-transform and a newly developed nonlinear mapping are used to obtain a structure-preserving mapping of high-dimensional data to a plane, suggesting that the potential power of nonlinear mappings is needed only for artificial data.

Journal ArticleDOI
TL;DR: The so-called posterior probability estimator, e, formed by averaging the minimum of the posterior probabilities over a set of initial or additional observations (which need not be classified) is considered in the context of estimating the overall actual error rate for the linear discriminant function appropriate for two multivariate normal populations with a common covariance matrix.

Journal ArticleDOI
Shin-ichi Hanaki1, Takemi Yamazaki1
TL;DR: Two symbol string recognition methods are proposed and investigated; the direct matching recognition (DMR) method and the unit structure recognition (USR) method, which worked efficiently for multi-stroke characters and greatly reduced dictionary update labor.

Journal ArticleDOI
TL;DR: This paper investigates the effect of intraclass correlation among training samples upon the misclassification probabilities of Bayes' procedure whenever the training samples are assumed to be serially correlated.

Journal ArticleDOI
Nobuyuki Goto1, Takashi Kondo1
TL;DR: Sufficient defect detection capability, as well as high inspection throughput, has been accomplished by hardware-implementation of the algorithm and inspection result utilization through retouching machines.