Pattern Recognition : Concepts, Methods, and Applications

by Sa, J. P. Marques De

• ISBN13: 9783540422976
• ISBN10: 3540422978
• Format: Hardcover
• Copyright: 2001-08-01
• Publisher: Springer Verlag

Summary

The book provides a comprehensive view of Pattern Recognition concepts and methods, illustrated with real-life applications in several areas. It is appropriate as a textbook of Pattern Recognition courses and also for professionals and researchers who need to apply Pattern Recognition techniques. These are explained in a unified an innovative way, with multiple examples enhacing the clarification of concepts and the application of methods. Recent methods and results in Pattern Recognition are also presented in a clear way. A CD-ROM offered with the book includes datasets and software tools, making it easier for the reader to follow the taught matters in a hands-on fashion right from the start.

Table of Contents

Preface	vii
Contents	xi
Symbols and Abbreviations	xvii
Basic Notions	1	(20)
Object Recognition	1	(1)
Pattern Similarity and PR Tasks	2	(7)
Classification Tasks	3	(3)
Regression Tasks	6	(2)
Description Tasks	8	(1)
Classes, Patterns and Features	9	(4)
PR Approaches	13	(3)
Data Clustering	14	(1)
Statistical Classification	14	(1)
Neural Networks	15	(1)
Structural PR	16	(1)
PR Project	16	(5)
Project Tasks	16	(2)
Training and Testing	18	(1)
PR Software	18	(2)
Bibliography	20	(1)
Pattern Discrimination	21	(32)
Decision Regions and Functions	21	(8)
Generalized Decision Functions	23	(3)
Hyperplane Separability	26	(3)
Feature Space Metrics	29	(4)
The Covariance Matrix	33	(6)
Principal Components	39	(2)
Feature Assessment	41	(5)
Graphic Inspection	42	(1)
Distribution Model Assessment	43	(1)
Statistical Inference Tests	44	(2)
The Dimensionality Ratio Problem	46	(7)
Bibliography	49	(1)
Exercises	49	(4)
Data Clustering	53	(26)
Unsupervised Classification	53	(2)
The Standardization Issue	55	(3)
Tree Clustering	58	(7)
Linkage Rules	60	(3)
Tree Clustering Experiments	63	(2)
Dimensional Reduction	65	(5)
K-Means Clustering	70	(3)
Cluster Validation	73	(6)
Bibliography	76	(1)
Exercises	77	(2)
Statistical Classification	79	(68)
Linear Discriminants	79	(11)
Minimum Distance Classifier	79	(3)
Euclidian Linear Discriminants	82	(3)
Mahalanobis Linear Discriminants	85	(3)
Fisher's Linear Discriminant	88	(2)
Bayesian Classification	90	(18)
Bayes Rule for Minimum Risk	90	(7)
Normal Bayesian Classification	97	(6)
Reject Region	103	(2)
Dimensionality Ratio and Error Estimation	105	(3)
Model-Free Techniques	108	(13)
The Parzen Window Method	110	(3)
The K-Nearest Neighbours Method	113	(3)
The ROC Curve	116	(5)
Feature Selection	121	(5)
Classifier Evaluation	126	(4)
Tree Classifiers	130	(8)
Decision Trees and Tables	130	(6)
Automatic Generation of Tree Classifiers	136	(2)
Statistical Classifiers in Data Mining	138	(9)
Bibliography	140	(2)
Exercises	142	(5)
Neural Networks	147	(96)
LMS Adjusted Discriminants	147	(8)
Activation Functions	155	(4)
The Percepton Concept	159	(8)
Neural Network Types	167	(4)
Multi-Layer Perceptrons	171	(13)
The Back-Propagation Algorithm	172	(3)
Practical aspects	175	(6)
Time Series	181	(3)
Performance of Neural Networks	184	(17)
Error Measures	184	(2)
The Hessian Matrix	186	(3)
Bias and Variance in NN Design	189	(3)
Network Complexity	192	(7)
Risk Minimization	199	(2)
Approximation Methods in NN Training	201	(6)
The Conjugate-Gradient Method	202	(3)
The Levenberg-Marquardt Method	205	(2)
Genetic Algorithms in NN Training	207	(5)
Radial Basis Functions	212	(3)
Support Vector Machines	215	(8)
Kohonen Networks	223	(3)
Hopfield Networks	226	(5)
Modular Neural Networks	231	(4)
Neural Networks in Data Mining	235	(8)
Bibliography	237	(2)
Exercises	239	(4)
Structural Pattern Recognition	243	(48)
Pattern Primitives	243	(4)
Signal Primitives	243	(2)
Image Primitives	245	(2)
Structural Representations	247	(3)
Strings	247	(1)
Graphs	248	(1)
Trees	249	(1)
Syntactic Analysis	250	(15)
String Grammars	250	(3)
Picture Description Language	253	(2)
Grammar Types	255	(2)
Finite-State Automata	257	(3)
Attributed Grammars	260	(1)
Stochastic Grammars	261	(3)
Grammatical Inference	264	(1)
Structural Matching	265	(26)
String Matching	265	(6)
Probablistic Relaxation Matching	271	(3)
Discrete Relaxation Matching	274	(1)
Relaxation Using Hopfield Networks	275	(4)
Graph and Tree Matching	279	(4)
Bibliography	283	(2)
Exercises	285	(6)
Appendix A - CD Datasets	291	(10)
A.1 Breat Tissue	291	(1)
A.2 Clusters	292	(1)
A.3 Cork Stoppers	292	(1)
A.4 Crimes	293	(1)
A.5 Cardiotocographic Data	293	(1)
A.6 Electrocardiograms	294	(1)
A.7 Foetal Heart Rate Signals	295	(1)
A.8 FHR-Apgar	295	(1)
A.9 Firms	296	(1)
A.10 Foetal Weight	296	(1)
A.11 Food	297	(1)
A.12 Fruits	297	(1)
A.13 Impulses on Noise	297	(1)
A.14 MLP Sets	298	(1)
A.15 Norm2c2d	298	(1)
A.16 Rocks	299	(1)
A.17 Stock Exchange	299	(1)
A.18 Tanks	300	(1)
A.19 Weather	300	(1)
Appendix B - CD Tools	301	(10)
B.1 Adaptive Filtering	301	(1)
B.2 Density Estimation	301	(1)
B.3 Design Set Size	302	(1)
B.4 Error Energy	303	(1)
B.5 Genetic Neural Networks	304	(2)
B.6 Hopfield network	306	(2)
B.7 k-NN Bounds	308	(1)
B.8 k-NN Classification	308	(1)
B.9 Perceptron	309	(1)
B.10 Syntactic Analysis	309	(2)
Appendix C - Orthonormal Transformation	311	(4)
Index	315