书名:
Introduction to Statistical Pattern Recognition
Keinosuke F'ukunaga
大小:616页
格式:pdf
目录:
Contents
Preface ............................................. xi
Acknowledgments .................................. xm
Chapter 1 Introduction 1
Formulation of Pattern Recognition Problems ......... 1
Process of Classifier Design ........................ 7
...
1.1
1.2
Notation ........................................ 9
References ..................................... 10
Chapter2 Random Vectors and Their Properties
Random Vectors and Their Distributions ............. 11
Estimation of Parameters ......................... 17
2.3 Linear Transformation ........................... 24
Computer Projects ............................... 47
Problems ....................................... 48
11
2.1
2.2
2.4 Various Properties of Eigenvalues and
Eigenvectors ................................... 35
References ..................................... 50
Vii
Viii Contents
Chapter 3 Hypothesis Testing 51
3.1
3.2
3.3
3.4
3.5 Sequential Hypothesis Testing .................... 110
Problems ...................................... 120
References .................................... 122
Hypothesis Tests for Two Classes ................... 51
Other Hypothesis Tests ........................... 65
Error Probability in Hypothesis Testing ............. 85
Upper Bounds on the Bayes Error .................. 97
Computer Projects .............................. 119
Chapter 4 Parametric Classifiers 124
4.1 The Bayes Linear Classifier ....................... 125
4.2 Linear Classifier Design ......................... 131
4.3 Quadratic Classifier Design ...................... 153
4.4 Other Classifiers ................................ 169
Computer Projects .............................. 176
Problems ...................................... 177
References ..................................... 180
Chapter 5 Parameter Estimation 181
5.1 Effect of Sample Size in Estimation ................ 182
5.2 Estimation of Classification Errors ................ 196
5.3 Holdout. LeaveOneOut. and Resubstitution
Methods ...................................... 219
5.4 Bootstrap Methods ............................. 238
Computer Projects .............................. 250
Problems ...................................... 250
References .................................... 252
Chapter 6 Nonparametric Density Estimation 254
6.1
6.2
6.3
Parzen Density Estimate ........................ 255
kNearest Neighbor Density Estimate .............. 268
Expansion by Basis Functions .................... 287
Computer Projects .............................. 295
Problems ..................................... 296
References .................................... 297
Contents ix
Chapter 7 Nonparametric Classification and
Error Estimation 300
7.1 General Discussion .............................. 301
7.2 Voting kNN Procedure - Asymptotic Analysis ...... 305
7.3 Voting kNN Procedure - Finite Sample Analysis ..... 313
7.4 Error Estimation ............................... 322
7.5 Miscellaneous Topics in the kNN Approach .......... 351
Computer Projects .............................. 362
Problems ...................................... 363
References ..................................... 364
Chapter 8 Successive Parameter Estimation 367
8.1 Successive Adjustment of a Linear Classifier ........ 367
8.2 Stochastic Approximation ....................... 375
8.3 Successive Bayes Estimation ..................... 389
Computer Projects ............................ 395
Problems .................................... 396
References ................................... 397
Chapter 9 Feature Extraction and Linear Mapping
9.1 The Discrete Karhunen-Lokve Expansion ........... 400
9.2 The Karhunen-LoBve Expansion for Random
Processes ..................................... 417
9.3
for Signal Representation 399
Estimation of Eigenvalues and Eigenvectors . . . . . . . . 425
Computer Projects .............................. 435
Problems ..................................... 438
References .................................... 440
Chapter 10 Feature Extraction and Linear Mapping
for Classification 441
10.1 General Problem Formulation .................... 442
10.2 Discriminant Analysis ......................... 445
10.3 Generalized Criteria ............................ 460
10.4 Nonparametric Discriminant Analysis . . . . . . . . . . . . 466
10.5 Sequential Selection of Quadratic Features . . . . . . . . . 480
10.6 Feature Subset Selection ........................ 489
X Contents
Computer Projects ............................. 503
Problems ..................................... 504
References .................................... 506
Chapter 11 Clustering 508
11.1 Parametric Clustering .......................... 509
11.2 Nonparametric Clustering ....................... 533
11.3 Selection of Representatives ..................... 549
Computer Projects ............................. 559
Problems ..................................... 560
References .................................... 562
Appendix A DERIVATIVES OF MATRICES ............. 564
Appendix B MATHEMATICAL FORMULAS ............ 572
Appendix C NORMAL ERROR TABLE .................5 76
Appendix D GAMMA FUNCTION TABLE .............. 578
Index ................................................ 579