Chang C.-I. (ed.) Hyperspectral Data Exploitation. Theory and Applications

John Wiley, 2007. — 443 p.Hyperspectral imaging has become one of most promising and emerging techniques in remote sensing. It has made great advance in recent years due to introduction of new techniques from versatile disciplines, particularly statistical signal processing from engineering aspects. Such clear evidence can be witnessed by hundreds of articles published in journals and conference proceedings every year as well as many annual conferences held in various venues. The rapid growth in this subject has made many researchers difficult to keep up with new developments and advances in its technology. Despite the fact that many books have been published in the area of remote sensing image processing, most have been focused on multispectral image processing rather than hyperspectral signal/image processing. Until recently, only a few appeared as book forms in this particular field. One example is my first book, Hyperspectral Imaging: Spectral Techniques for Detection and Classification, published in 2003 by Kluwer/Plenum Academic Publishers (now part of Springer-Verlag Publishers); this was primarily written for subpixel detection and mixed pixel classification that were designed and developed in my laboratory.
Unfortunately, many other topics are also of interest, but were not covered in this particular book. In order to address this need, I have made a significant effort to invite experts in hyperspectral imaging from academia and industries to write chapters in their expertise and share their research works with readers. This book is essentially a result of their contributions. A total of 13 chapters (Chapters 2 to 14) are included in this book and cover a wide spectrum of topics in hyperspectral data exploitation including imaging systems, data modeling, data representation, band selection and partition, and classification to data compression. Each chapter has been contributed by an expert in his/her specialty or by experts in their specialties. Also included is Chapter 1, an overview written by me which provides readers with a discussion on design philosophy in developing hyperspectral imaging techniques from a hyperspectral imagery point of view as well as brief reviews of each of the 13 chapters including coherent connections among different chapters. Therefore, this chapter can serve as a guide to direct readers to particular topics in which they are interested.
The ultimate goal of this book is to offer readers a peek at the cutting-edge research in hyperspectral data exploitation. In particular, this book can be found very useful for practitioners and engineers who are interested in this area. It is hoped that the chapters presented in this book have just done that.Overview
Tutorials
Hyperspectral Imaging Systems
Information-Processed Matched Filters for Hyperspectral Target Detection and Classification
Theory
An Optical Real-Time Adaptive Spectral Identification System (ORASIS)
Stochastic Mixture Modeling
Unmixing Hyperspectral Data: Independent and Dependent Component Analysis
Maximum Volume Transform for Endmember Spectra Determination
Hyperspectral Data Representation
Optimal Band Selection and Utility Evaluation for Spectral Systems
Feature Reduction for Classification Purpose
Semisupervised Support Vector Machines for Classification of Hyperspectral Remote Sensing Images
Applications
Decision Fusion for Hyperspectral Classification
Morphological Hyperspectral Image Classification: A Parallel Processing Perspective
Three-Dimensional Wavelet-Based Compression of Hyperspectral Imagery