Some approaches to pattern recognition problems

A new approach to selecting the Gibbs distribution in models of objects to be recognized is proposed. This approach proposes to determine the lower and upper bounds for probabilities of the object under study. The distance between these bounds may be used as a measure of error in pattern recognition problems. __________ Translated from Kibernetika i Sistemnyi Analiz, No. 6, pp. 55–69, November–December 2007.     

A training sample sequence planning method for pattern recognition problems

In solving pattern recognition problems, many classification methods, such as the nearest-neighbor (NN) rule, need to determine prototypes from a training set. To improve the performance of these classifiers in finding an efficient set of prototypes, this paper introduces a training sample sequence planning method. In particular, by estimating the relative nearness of the training samples to the decision boundary, the approach proposed here incrementally increases the number of prototypes until the desired classification accuracy has been reached. This approach has been tested with a NN classification method and a neural network training approach. Studies based on both artificial and real data demonstrate that higher classification accuracy can be achieved with fewer prototypes.     

Efficient decision tree design for discrete variable pattern recognition problems

An algorithm is developed for the design of an efficient decision tree with application to the pattern recognition problems involving discrete variables. The problem of evaluating an extremely large number of trees in search of a minimum cost decision tree is tackled by defining a criterion to estimate the minimum expected cost of a tree in terms of the weights of its terminal nodes and costs of the measurements, which then is used to establish the search procedure for the efficient decision tree. The concept of prime events is used to obtain the number of modes and the corresponding weights in the design samples. An application of the proposed algorithm is presented for the design of an efficient decision tree for classifying Devanagri numerals.     

Potential functions-based wavelet transforms in pattern recognition problems

The problem of finding a given function in the composition of another function for the one-dimensional case, as well as the problem of finding particular fragments in the composition of a given surface, are considered. The application of wavelet transforms in solving the posed problems is shown by particular examples.     

A fuzzy ARTMAP nonparametric probability estimator fornonstationary pattern recognition problems

An incremental, nonparametric probability estimation procedure using the fuzzy ARTMAP (adaptive resonance theory-supervised predictive mapping) neural network is introduced. In the slow-learning mode, fuzzy ARTMAP searches for patterns of data on which to build ever more accurate estimates. In max-nodes mode, the network initially learns a fixed number of categories, and weights are then adjusted gradually.     

Some methods for pattern recognition

The present study is dedicated to pattern recognition of alphabetical and numerical symbols. A recognition method based on modification of Hausdorff metrics is presented; then, another method named the method of radial neighborhoods is described. A series of experiments on test patterns are conducted.     

Image normalization for pattern recognition

In general, there are four basic forms of distortion in the recognition of planar patterns: translation, rotation, scaling and skew. In this paper, a normalization algorithm has been developed which transforms pattern into its normal form such that it is invariant to translation, rotation, scaling and skew. After normalization, the recognition can be performed by a simple matching method. In the algorithm, we first compute the covariance matrix of a given pattern. Then we rotate the pattern according to the eigenvectors of the covariance matrix, and scale the pattern along the two eigenvectors according to the eigenvalues to bring the pattern to its most compact form. After the process, the pattern is invariant to translation, scaling and skew. Only the rotation problem remains unsolved. By applying the tensor theory, we find a rotation angle which can make the pattern invariant to rotation. Thus, the resulting pattern is invariant to translation, rotation, scaling and skew. The planar image used in this algorithm may be curved, shaped, a grey-level image or a coloured image, so its applications are wide, including recognition problems about curve, shape, grey-level and coloured patterns. The technique suggested in this paper is easy, does not need much computation, and can serve as a pre-processing step in computer vision applications.     

Artificial data for pattern recognition

The paper describes the generation of three types of artificial data and their use as test material in pattern recognition research. Type A data: The user defines the perfect decision surface. The classes are separable and the pdf's flat. This type is useful in two ways: (i) To investigate whether a learning procedure can achieve a minimal-cost solution. (ii) To compare the powers of two classifiers. Type B data: The user defines the optimal decision surface. The classes are not separable; the degree of overlap between the classes can be controlled by the user. The pdf's are approximately flat, except in regions close to this optimal decision boundary. This type is useful in the following ways: (i) To study the effect of varying the overlap between classes upon a learning procedure. (ii) To compare the powers of two classifiers on a random problem. Type C data: This type is a model of natural, clustered data. The user specifies the location, height, and spread of a number of “hills” in the pdf (for each class). These parameters allow us to calculate the pdf's and hence the Bayes' classification, at any given point. This provides a powerful tool for the objective evaluation of a learning classifier, operating on a realistic problem.     

Moment invariants for pattern recognition

Invariant combinations of moments of arbitrary order are defined. Application to a vehicle image shows that a reconstructed image having <10% error may be obtained by using invariants formed from moments uop to order eight.     

Syntactic pattern recognition for the recognition of bright spots

Syntactic pattern recognition techniques are applied to the analysis of one-dimensional seismic traces and two-dimensional seismograms for the detection of bright spots. The calculation between error probability and Levenshtein distance is proposed. The system for two-dimensional seismic analysis includes three kinds of string distance computation to test the continuity of a bright spot pattern.     

Adaptive correlation filters for pattern recognition

Adaptive correlation filters based on synthetic discriminant functions (SDFs) for reliable pattern recognition are proposed. A given value of discrimination capability can be achieved by adapting a SDF filter to the input scene. This can be done by iterative training. Computer simulation results obtained with the proposed filters are compared with those of various correlation filters in terms of recognition performance. The text was submitted by the authors in English. Vitaly Kober obtained his MS degree in Applied Mathematics from the Air-Space University of Samara (Russia) in 1984 and his PhD degree in 1992 and Doctor of Sciences degree in 2004 in Image Processing from the Institute of Information Transmission Problems, Russian Academy of Sciences. He is now a titular researcher at the Centro de Investigatión Cientifica y de Educatión Superior de Ensenada (Cicese), Mexico. His research interests include signal and image processing and pattern recognition. Mikhail Mozerov received his MS degree in Physics from Moscow State University in 1982 and his PhD degree in Image Processing from the Institute of Information Transmission Problems, Russian Academy of Sciences, in 1995. He is with the Laboratory of Digital Optics of the Institute of Information Transmission Problems, Russian Academy of Sciences. His research interests include signal and image processing, pattern recognition, and digital holography. Iosif A. Ovseyevich graduated from the Moscow Electrotechnical Institute of Telecommunications. He received his candidate’s degree in 1953 and doctoral degree in Information Theory in 1972. At present he is Emeritus Professor at the Institute of Information Transmission Problems, Russian Academy of Sciences. His research interests include information theory, signal processing, and expert systems. He is a Member of the IEEE and Popov Radio Society.     

Visible models for interactive pattern recognition

The exchange of information between human and machine has been a bottleneck in interactive visual classification. The visible model of an object to be recognized is an abstraction of the object superimposed on its picture. It is constructed by the machine but it can be modified by the operator. The model guides the extraction of features from the picture. The classes are rank ordered according to the similarities (in the hidden high-dimensional feature space) between the unknown picture and a set of labeled reference pictures. The operator can either accept one of the top three candidates by clicking on a displayed reference picture, or modify the model. Model adjustment results in the extraction of new features, and a new rank ordering. The model and feature extraction parameters are re-estimated after each classified object, with its model and label, is added to the reference database. Pilot experiments show that interactive recognition of flowers and faces is more accurate than automated classification, faster than unaided human classification, and that both machine and human performance improve with use.     

Structured neural networks for pattern recognition

This paper proposes a novel approach for the design of structures of neural networks for pattern recognition. The basic idea lies in subdividing the whole classification problem in smaller and simpler problems at different levels, each managed by appropriate components of a complex neural architecture. Three neural structures are presented and applied in a surveillance system aimed at monitoring a railway waiting room classifying potential dangerous situations. Each architecture is composed by nodes, which are actual multilayer perceptrons trained to discriminate between subsets of classes until a complete separation among the classes is achieved. This approach showed better performances with respect to a classical statistical classification procedures and to a single neural network.     

Using interpoint distances for pattern recognition

The simplest problem of pattern recognition is considered, viz., recognition of two classes of objects given by points in Euclidean space; the recognition is fulfilled using the fraction of samples that fall within a ball of radius r with the center at the specified point. The interpoint distances are considered both within the totalities and between them. Their pairwise asymptotic independence is proved, which allows reducing the multidimensional distribution to one-dimensional, where simple yet important problems can be solved. For instance, one can check the significance of the difference between the class distributions, choose the optimal value of radius r, and stably estimate the error of recognition. Aleksandr Mikhailovich Shurygin. Born 1931. Graduated from Moscow State University, Faculty of Geology (1949–1954) and Faculty of Mechanics and Mathematics (1960–1965). Received candidate’s degree in Geology and Mineralogy (Tectonics) in 1959 and doctoral degree (Information Theory) in 2002. Since 1970, he has had the title of Senior Researcher. He works as a leading researcher at the Faculty of Computational Mathematics and Cybernetics of MSU and as a senior researcher at the Scientific Council of Cybernetics of the RAS. Author of more than 130 scientific publications, including five monographs.