Pattern Recognition Framework Research Articles

Classification methods for random utility models with i.i.d. disturbances under the most probable alternative rule

In this paper random utility maximization based on maximization of correct classification of the choice decisions over a given data set is considered. It is shown that if the disturbance vector in the random utility model is independent and identically distributed, then preference determination based on the most probable alternative reduces to deterministic utility maximization. As a consequence of the above equivalence, the form of the error distribution (normal, Weibull, uniform etc.) plays no role in the determination of the preferred alternative. Parameter estimation under the most probable alternative rule is carried out using two methods. The first is based on the solution of an appropriately defined system of linear inequalities and the second one is based on the function optimization of a newly proposed function, whose optimum is achieved when the number of correctly classified individuals is maximized. The ability to use these algorithms in the framework of pattern recognition and machine learning is pointed out. Simulations and a real case study involving intercity travel behavior are employed to assess the proposed methods.

Capitalize on dimensionality increasing techniques for improving face recognition grand challenge performance

This paper presents a novel pattern recognition framework by capitalizing on dimensionality increasing techniques. In particular, the framework integrates Gabor image representation, a novel multiclass Kernel Fisher Analysis (KFA) method, and fractional power polynomial models for improving pattern recognition performance. Gabor image representation, which increases dimensionality by incorporating Gabor filters with different scales and orientations, is characterized by spatial frequency, spatial locality, and orientational selectivity for coping with image variabilities such as illumination variations. The KFA method first performs nonlinear mapping from the input space to a high-dimensional feature space, and then implements the multiclass Fisher discriminant analysis in the feature space. The significance of the nonlinear mapping is that it increases the discriminating power of the KFA method, which is linear in the feature space but nonlinear in the input space. The novelty of the KFA method comes from the fact that 1) it extends the two-class kernel Fisher methods by addressing multiclass pattern classification problems and 2) it improves upon the traditional Generalized Discriminant Analysis (GDA) method by deriving a unique solution (compared to the GDA solution, which is not unique). The fractional power polynomial models further improve performance of the proposed pattern recognition framework. Experiments on face recognition using both the FERET database and the FRGC (Face Recognition Grand Challenge) databases show the feasibility of the proposed framework. In particular, experimental results using the FERET database show that the KFA method performs better than the GDA method and the fractional power polynomial models help both the KFA method and the GDA method improve their face recognition performance. Experimental results using the FRGC databases show that the proposed pattern recognition framework improves face recognition performance upon the BEE baseline algorithm and the LDA-based baseline algorithm by large margins.

An intelligent system for automatic detection of gastrointestinal adenomas in video endoscopy

Today 95% of all gastrointestinal carcinomas are believed to arise from adenomas. The early detection of adenomas could prevent their evolution to cancer. A novel system for the support of the detection of adenomas in gastrointestinal video endoscopy is presented. Unlike other systems, it accepts standard low-resolution video input thus requiring less computational resources and facilitating both portability and the potential to be used in telemedicine applications. It combines intelligent processing techniques of SVMs and color–texture analysis methodologies into a sound pattern recognition framework. Concerning the system's accuracy this was measured using ROC analysis and found to exceed 94%.

Conception of complex probabilistic neural network system for classification of partial discharge patterns using multifarious inputs

Pattern recognition has a long history within electrical engineering but has recently become much more widespread as the automated capture of signal and images has been cheaper. Very many of the application of neural networks are to classification, and so are within the field of pattern recognition and classification. In this paper, we explore how probabilistic neural networks fit into the earlier framework of pattern recognition of partial discharge patterns since the PD patterns are an important tool for diagnosis of HV insulation systems. Skilled humans can identify the possible insulation defects in various representations of partial discharge (PD) data. One of the most widely used representation is phase resolved PD (PRPD) patterns. Also this paper describes a method for the automated recognition of PRPD patterns using a novel complex probabilistic neural network system for the actual classification task. The efficacy of composite neural network developed using probabilistic neural network is examined.

Multidimensional Protein Identification Technology (MudPIT): Technical Overview of a Profiling Method Optimized for the Comprehensive Proteomic Investigation of Normal and Diseased Heart Tissue

An optimized analytical expression profiling strategy based on gel-free multidimensional protein identification technology (MudPIT) is reported for the systematic investigation of biochemical (mal)-adaptations associated with healthy and diseased heart tissue. Enhanced shotgun proteomic detection coverage and improved biological inference is achieved by pre-fractionation of excised mouse cardiac muscle into subcellular components, with each organellar fraction investigated exhaustively using multiple repeat MudPIT analyses. Functional-enrichment, high-confidence identification, and relative quantification of hundreds of organelle- and tissue-specific proteins are achieved readily, including detection of low abundance transcriptional regulators, signaling factors, and proteins linked to cardiac disease. Important technical issues relating to data validation, including minimization of artifacts stemming from biased under-sampling and spurious false discovery, together with suggestions for further fine-tuning of sample preparation, are discussed. A framework for follow-up bioinformatic examination, pattern recognition, and data mining is also presented in the context of a stringent application of MudPIT for probing fundamental aspects of heart muscle physiology as well as the discovery of perturbations associated with heart failure.

Relevance vector machine for optical diagnosis of cancer

A probability-based, robust diagnostic algorithm is an essential requirement for successful clinical use of optical spectroscopy for cancer diagnosis. This study reports the use of the theory of relevance vector machine (RVM), a recent Bayesian machine-learning framework of statistical pattern recognition, for development of a fully probabilistic algorithm for autofluorescence diagnosis of early stage cancer of human oral cavity. It also presents a comparative evaluation of the diagnostic efficacy of the RVM algorithm with that based on support vector machine (SVM) that has recently received considerable attention for this purpose. The diagnostic algorithms were developed using in vivo autofluorescence spectral data acquired from human oral cavity with a N(2) laser-based portable fluorimeter. The spectral data of both patients as well as normal volunteers, enrolled at Out Patient department of the Govt. Cancer Hospital, Indore for screening of oral cavity, were used for this purpose. The patients selected had no prior confirmed malignancy and were diagnosed of squamous cell carcinoma (SCC), Grade-I on the basis of histopathology of biopsy taken from abnormal site subsequent to acquisition of spectra. Autofluorescence spectra were recorded from a total of 171 tissue sites from 16 patients and 154 healthy squamous tissue sites from 13 normal volunteers. Of 171 tissues sites from patients, 83 were SCC and the rest were contralateral uninvolved squamous tissue. Each site was treated separately and classified via the diagnostic algorithm developed. Instead of the spectral data from uninvolved sites of patients, the data from normal volunteers were used as the normal database for the development of diagnostic algorithms. The diagnostic algorithms based on RVM were found to provide classification performance comparable to the state-of-the-art SVMs, while at the same time explicitly predicting the probability of class membership. The sensitivity and specificity towards cancer were up to 88% and 95% for the training set data based on leave- one-out cross validation and up to 91% and 96% for the validation set data. When implemented on the spectral data of the uninvolved oral cavity sites from the patients, it yielded a specificity of up to 91%. The Bayesian framework of RVM formulation makes it possible to predict the posterior probability of class membership in discriminating early SCC from the normal squamous tissue sites of the oral cavity in contrast to dichotomous classification provided by the non-Bayesian SVM. Such classification is very helpful in handling asymmetric misclassification costs like assigning different weights for having a false negative result for identifying cancer compared to false positive. The results further demonstrate that for comparable diagnostic performances, the RVM-based algorithms use significantly fewer kernel functions and do not need to estimate any hoc parameters associated with the learning or the optimization technique to be used. This implies a considerable saving in memory and computation in a practical implementation.

A large deviation approach to normality testing

A new approach to the classical problem of determining whether or not a set of data has been sampled from a univariate normal distribution is considered. The problem is posed in a pattern recognition framework and the concept of an asymptotically optimal expert is introduced. Some results of Large Deviation theory are used in the selection of experts so that their combined abilities lead to a suitable discrimination procedure. The fact that good asymptotical properties can be extended to finite samples with small sizes is the basis for the rest of the work. The performance of the proposed procedure is also compared to that of some known normality tests. An additional advantage of the procedure is the possibility of computing a reliable measure of support such as the Bayesian posterior probabilities as opposed to the P -values. Furthermore, some extra information can be obtained concerning the type of deviation from normality which is present in the data.

PRISM ? A novel framework for pattern recognition

In this paper, we introduce a new model of solving pattern recognition tasks called PRISM (Pattern Recognition using Information Slicing Method). The main concept behind PRISM is the slicing of information through multiple planes across different feature axes to generate a number of cells. The number of cells created and their volume depends upon the number partitions per axes. In this context we define resolution as the number of partitions per axes. In this paper, we make the following contributions. First, we provide a brief survey of the class separability measures and feature partitioning schemes used for pattern recognition. Secondly, we define the PRISM framework and the algorithm for data assignment to cells. Thirdly, we detail four important concepts in PRISM: purity, neighbourhood separability, collective entropy, and data compactness. The first two measures define the data complexity, the next measure relates to uncertainty, and the last measure defines the alternative to statistical data variance in the PRISM framework. Fourthly, we investigate the variability in the estimates of these measures depending on the placement of partitions on each feature axis. Finally, we give an overview of experimental successes achieved with PRISM in the areas of classification complexity estimation and feature selection.

Hidden Markov model-based real-time transient identifications in nuclear power plants

In this article, a transient identification method based on a stochastic approach with the hidden Markov model (HMM) has been suggested and evaluated experimentally for the classification of nine types of transients in nuclear power plants (NPPs). A transient is defined as when a plant proceeds to an abnormal state from a normal state. Identification of the types of transients during an early accident stage in NPPs is crucial for proper action selection. The transient can be identified by its unique time-dependent patterns related to the principal variables. The HMM, a double-stochastic process, can be applied to transient identification that is a spatial and temporal classification problem under a statistical pattern-recognition framework. The trained HMM is created for each transient from a set of training data by the maximum-likelihood estimation method which uses a forward-backward algorithm and the Baum-Welch re-estimation algorithm. The transient identification is determined by calculating which model has the highest probability for given test data using the Viterbi algorithm. Several experimental tests have been performed with normalization methods, clustering algorithms, and a number of states in HMM. There are also a few experimental tests that have been performed, including superimposing random noise, adding systematic error, and adding untrained transients to verify its performance and robustness. The proposed real-time transient identification system has been proven to have many advantages, although there are still some problems that should be solved before applying it to an operating NPP. Further efforts are being made to improve the system performance and robustness in order to demonstrate reliability and accuracy to the required level. © 2002 Wiley Periodicals, Inc.

Multiple Classifier Fusion in Probabilistic Neural Networks

The main motivation of this paper is to design a statistically well justified and biologically compatible neural network model and, in particular, to suggest a theoretical interpretation of the well known high parallelism of biological neural networks. We consider a novel probabilistic approach to neural networks developed in the framework of statistical pattern recognition, and refer to a series of theoretical results published earlier. It is shown that the proposed parallel fusion of probabilistic neural networks produces biologically plausible structures and improves the resulting recognition performance. The complete design methodology based on the EM algorithm has been applied to recognise unconstrained handwritten numerals from the database of Concordia University Montreal. We achieved a recognition accuracy of about 95%, which is comparable with other published results.

A Three-Dimensional Uniquely Parsable Array Grammar that Generates and Parses Cubes

A uniquely parsable array grammar (UPAG) introduced by Yamamoto and Morita is a subclass of isometric array grammar (IAG) in which parsing can be performed without backtracking. Hence, we can use a UPAG as an efficient two-dimensional pattern recognition mechanism, if the pattern set is properly described by a UPAG. Furthermore, since a UPAG admits parallel parsing, it can also be used as a formal framework for parallel pattern recognition. In this paper, we show UPAG is also useful to generate and recognize three-dimensional patterns. We construct a UPAG that generate any size of cubes. This UPAG can recognize cubes in linear time of the length of its side by maximum parallel reduction. To construct the grammar, we have made a tool for designing three-dimensional IAG.

A conceptual development framework for intuitive human pattern recognition

In human visual perception, there is an intuitive tolerance (perceptual constancy) for perceiving raw stimulus patterns which may be rotated, scaled, deformed, or noisy to some extent for a learned pattern. Such an intuitive property of perception is a major feature of human pattern recognition, where a recognition mechanism (consider the character recognition for example) may not necessarily follow in a stroke‐based (feature‐oriented) scheme; rather, it may follow in a whole‐image‐based (spatial‐oriented) scheme. In this paper, we present a conceptual development framework for intuitive human pattern recognition. The implementation model is called I‐net and basically consists of three parts: an attention mechanism, a generalization mechanism, and a recognition mechanism. The paper begins with describing the intuitive properties involved in recognizing 2‐D binary patterns, following an introduction to intuitive human pattern recognition. Then the details of the I‐net are described and experimental results are presented. The paper concludes that the proposed framework provides another direction for approaching human pattern recognition.

Range Image Segmentation through Pattern Analysis of the Multiscale Wavelet Transform

This work presents an image segmentation method for range data that uses multiscale wavelet analysis in combination with statistical pattern recognition. A feature-detection framework based on multiscale analysis and pattern recognition has several potential advantages over other feature detection systems. These advantages are detection of features at different scales (i.e., features of all sizes), robustness, and few or no free parameters. Our system creates a fuzzy edge map and derives a segmentation from this edge detection. A scale-space signature is the vector of measurements at different scales taken at a single point in an image. We analyze these 1-D signatures with traditional pattern-recognition methods. We train a pattern-recognition system with scale-space signatures from the edge points of a training image. Once trained, the system determines the degree ofedgenessof points in a new image. The goal is to create a system that exploits the advantages of a multiscale, pattern-recognition framework.

On a Kernel-Based Method for Pattern Recognition, Regression, Approximation, and Operator Inversion

We present a kernel-based framework for pattern recognition, regression estimation, function approximation, and multiple operator inversion. Adopting a regularization-theoretic framework, the above are formulated as constrained optimization problems. Previous approaches such as ridge regression, support vector methods, and regularization networks are included as special cases. We show connections between the cost function and some properties up to now believed to apply to support vector machines only. For appropriately chosen cost functions, the optimal solution of all the problems described above can be found by solving a simple quadratic programming problem.

A Bayesian framework for deformable pattern recognition with application to handwritten character recognition

Deformable models have recently been proposed for many pattern recognition applications due to their ability to handle large shape variations. These proposed approaches represent patterns or shapes as deformable models, which deform themselves to match with the input image, and subsequently feed the extracted information into a classifier. The three components-modeling, matching, and classification-are often treated as independent tasks. In this paper, we study how to integrate deformable models into a Bayesian framework as a unified approach for modeling, matching, and classifying shapes. Handwritten character recognition serves as a testbed for evaluating the approach. With the use of our system, recognition is invariant to affine transformation as well as other handwriting variations. In addition, no preprocessing or manual setting of hyperparameters (e.g., regularization parameter and character width) is required. Besides, issues on the incorporation of constraints on model flexibility, detection of subparts, and speed-up are investigated. Using a model set with only 23 prototypes without any discriminative training, we can achieve an accuracy of 94.7 percent with no rejection on a subset (11,791 images by 100 writers) of handwritten digits from the NIST SD-1 dataset.

A Multiobjective Evolutionary Approach to Pattern Recognition for Robust Diagnosis of Process Faults

The problem of robust model-based diagnosis of process faults is addressed in the framework of pattern recognition. Evolutionary algorithms of genetic type are used to solve both problems of feature selection and classifier design by means of multiobjective optimization. Process coefficients are directly identified by an on-line procedure. Symptoms are then evaluated by a non-parametric classifier. Application to a laboratory process is included. A diagnosis subsystem is designed and implemented in real-time to detect incipient faults in the components of a three-tank system.

Bayesian Neural Networks

Bayesian techniques have been developed over many years in a range of different fields, but have only recently been applied to the problem of learning in neural networks. As well as providing a consistent framework for statistical pattern recognition, the Bayesian approach offers a number of practical advantages including a solution to the problem of over-fitting. This article provides an introductory overview of the application of Bayesian methods to neural networks. It assumes the reader is familiar with standard feed-forward network models and how to train them using conventional techniques

A unifying framework for invariant pattern recognition

We introduce a group-theoretic model of invariant pattern recognition, the Group Representation Network. We show that many standard invariance techniques can be viewed as GRNs, including the DFT power spectrum, higher order neural network and fast translation-invariant transform.

A comparison of connectionist models of music recognition and human performance

Current artificial neural network or connectionist models of music cognition embody feature-extraction and feature-weighting principles. This paper reports two experiments which seek evidence for similar processes mediating recognition of short musical compositions by musically trained and untrained listeners. The experiments are cast within a pattern recognition framework based on the vision-audition analogue wherein music is considered an auditory pattern consisting of local and global features. Local features such as inter-note interval, and global features such as melodic contour, are derived from a two-dimensional matrix in which music is represented as a series of frequencies plotted over time.

Classification of electrocardiographic signals: a fuzzy pattern matching approach

In this paper we will deal with the problem of classification and interpretation of electrocardiograms (ECG) in the presence of uncertain environment. The problem can be conveniently handled in a general framework of pattern recognition. Uncertainty that is evident there relates both to (i) a perception perspective established in the classification process as well as (ii) final class assignment. Due to this type of uncertainty which essentially derives from a lack of sharp boundaries (i.e. precise definitions) of the features of the signal and the partial class membership (that is characteristic for unclear, borderline cases) of individual signals a methodology of fuzzy set theory is appropriate. Within this framework it will be clarified how all perception properties concerning signal processing and its interpretation can be handled.