Зарегистрироваться
Восстановить пароль
FAQ по входу

Bow S.-T. Pattern Recognition and Image Preprocessing. Second Edition

  • Файл формата pdf
  • размером 11,13 МБ
  • Добавлен пользователем , дата добавления неизвестна
  • Отредактирован
Bow S.-T. Pattern Recognition and Image Preprocessing. Second Edition
Издательство Marcel Dekker, 2002, -690 pp.
This book is based in part on my earlier work, Pattern Recognition and Image Preprocessing, which was published in 1992 and reprinted in 1999. At the request of the publisher, in this expanded edition, I am including most of the supplementary materials added to my lectures from year to year since 1992 while I used this book as a text for two courses in pattern recognition and image processing.
Pattern recognition (or pattern classification) can be broadly defined as a process to generate a meaningful description of data and a deeper understanding of a problem through manipulation of a large set of primitive and quantifying data. The set inevitably includes image data—as a matter of fact, some of the data may come directly after the digitization of an actual natural scenic image. Some of that large data set may come from statistics, a document, or graphics, and is eventually expected to be in a visual form. Preprocessing of these data is necessary for error corrections, for image enhancement, and for their understanding and recognition. Preprocessing operations are generally classified as ‘‘low-level’’ operations, while pattern recognition including analysis, description, and understanding of the image (or the large data set), is high-level processing. The strategies and techniques chosen for the low- and high-level processing are interrelated and interdependent. Appropriate acquisition and preprocessing of the original data would alleviate the effort of pattern recognition to some extent. For a specific pattern recognition task, we frequently require a special method for the acquisition of data and its processing. For this reason, I have integrated these two levels of processing into a single book. Together with some exemplary paradigms, this book exposes readers to the whole process in the design of a good pattern recognition system and inspires them to seek applications within their own sphere of influence and personal experience.
Theory and applications are both important topics in the pattern recognition discussion. They are treated on a pragmatic basis in this book. We chose ‘‘application’’ as a vehicle through which to investigate many of the disciplines.
Recently, neural computing has been emerging as a practical technology with successful applications in many fields. The majority of these applications are concerned with problems in pattern recognition. Hence, in this edition we elaborate our discussion of neural networks for pattern recognition, with emphasis on multilayer perceptron, radial basis functions, the Hamming net, the Kohonen self-organizing feature map, and the Hopfield net. These five neural models are presented through simple examples to show the step-by-step procedure for neural computing to help readers start their computer implementation for more complex problems.
The wavelet is a good mathematical tool to extract the local features of variable sizes, variable frequencies, and variable locations in the image; it is very effective in the compression of image data. A new chapter on the wavelet and wavelet transform has been added in this edition. Some work done in our laboratory on wavelet tree-structure-based image compression, wavelet-based morphological processing for image noise reduction, and wavelet-based noise reduction for images with extremely high noise content is presented.
The materials collected for this book are grouped into five parts. Part I emphasizes the principles of decision theoretic pattern recognition. Part II introduces neural networks for pattern recognition. Part III deals with data preprocessing for pictorial pattern recognition. Part IV gives some current examples of applications to inspire readers and interest them in attacking realworld problems in their field with the pattern recognition technique and build their confidence in the capability and feasibility of this technique. Part V discusses some of the practical concerns in image preprocessing and pattern recognition.
Pattern Recognition
Introduction
Supervised and Unsupervised Learning in Pattern Recognition
Nonparametric Decision Theoretic Classification
Nonparametric (Distribution-Free) Training of Discriminant Functions
Statistical Discriminant Functions
Clustering Analysis and Unsupervised Learning
Dimensionality Reduction and Feature Selection
Neural Networks for Pattern Recognition
Multilayer Perceptron
Radial Basis Function Networks
Hamming Net and Kohonen Self-Organizing Feature Map
The Hopfield Model
Data Preprocessing for Pictorial Pattern Recognition
Preprocessing in the Spatial Domain
Pictorial Data Preprocessing and Shape Analysis
Transforms and Image Processing in the Transform Domain
Wavelets and Wavelet Transform
Applications
Exemplary Applications
Practical Concerns of Image Processing and Pattern Recognition
Computer System Architectures for Image Processing and Pattern Recognition
A: Digitized Images
B: Image Model and Discrete Mathematics
C: Digital Image Fundamentals
D: Matrix Manipulation
E: Eigenvectors and Eigenvalues of an Operator
F: Notation
  • Чтобы скачать этот файл зарегистрируйтесь и/или войдите на сайт используя форму сверху.
  • Регистрация