Introduction to Statistical Pattern Recognition

By Keinosuke Fukunaga

• PrefaceAcknowledgmentsChapter 1 Introduction1.1 Formulation of Pattern Recognition Problems1.2 Process of Classifier DesignNotationReferencesChapter 2 Random Vectors and Their Properties2.1 Random Vectors and Their Distributions2.2 Estimation of Parameters2.3 Linear Transformation2.4 Various Properties of Eigenvalues and EigenvectorsComputer ProjectsProblemsReferencesChapter 3 Hypothesis Testing3.1 Hypothesis Tests for Two Classes3.2 Other Hypothesis Tests3.3 Error Probability in Hypothesis Testing3.4 Upper Bounds on the Bayes Error3.5 Sequential Hypothesis TestingComputer ProjectsProblemsReferencesChapter 4 Parametric Classifiers4.1 The Bayes Linear Classifier4.2 Linear Classifier Design4.3 Quadratic Classifier Design4.4 Other ClassifiersComputer ProjectsProblemsReferencesChapter5 Parameter Estimation5.1 Effect of Sample Size in Estimation5.2 Estimation of Classification Errors5.3 Holdout, Leave-One-Out, and Resubstitution Methods5.4 Bootstrap MethodsComputer ProjectsProblemsReferencesChapter 6 Nonparametric Density Estimation6.1 Parzen Density Estimate6.2 kNearest Neighbor Density Estimate6.3 Expansion by Basis FunctionsComputer ProjectsProblemsReferencesChapter 7 Nonparametric Classification and Error Estimation7.1 General Discussion7.2 Voting kNN Procedure — Asymptotic Analysis7.3 Voting kNN Procedure — Finite Sample Analysis7.4 Error Estimation7.5 Miscellaneous Topics in the kNN ApproachComputer ProjectsProblemsReferencesChapter 8 Successive Parameter Estimation8.1 Successive Adjustment of a Linear Classifier8.2 Stochastic Approximation8.3 Successive Bayes EstimationComputer ProjectsProblemsReferencesChapter 9 Feature Extraction and Linear Mapping for Signal Representation9.1 The Discrete Karhunen-Loéve Expansion9.2 The Karhunen-Loéve Expansion for Random Processes9.3 Estimation of Eigenvalues and EigenvectorsComputer ProjectsProblemsReferencesChapter 10 Feature Extraction and Linear Mapping for Classification10.1 General Problem Formulation10.2 Discriminant Analysis10.3 Generalized Criteria10.4 Nonparametric Discriminant Analysis10.5 Sequential Selection of Quadratic Features10.6 Feature Subset SelectionComputer ProjectsProblemsReferencesChapter 11 Clustering11.1 Parametric Clustering11.2 Nonparametric Clustering11.3 Selection of RepresentativesComputer ProjectsProblemsReferencesAppendix A Derivatives of MatricesAppendix B Mathematical FormulasAppendix C Normal Error TableAppendix D Gamma Function TableIndex