Издательство Springer, 2013, -199 pp.Hyperspectral imaging is a fairly recent technique that has an excellent potential for applications in many diverse areas including the primary area of remote sensing. A fine spectral sampling allows us to capture the spectral signature with great details. A huge amount of data to represent a single scene, however, imposes difficulties in visualizing the image. This motivated us to develop various techniques for visualizing the hyperspectral data cube. Pixel-level image fusion appeared to be a natural choice for visualization of the entire data cube. While working on the visualization problem, it was felt that the available literature is quite scattered and very sparse compared to the much matured area of basic image fusion. Hence we embarked on writing a monograph that will serve as a reference book in this area to all students and practitioners of hyperspectral imaging technology. The monograph serves two purposes—it provides an up-todate reference to all research work, and it also provides implementation details of various visualization techniques with the hope that one should be able to implement them easily. We also present a comparative analysis of various competing techniques to enable researchers to decide which method would be more suitable to them. As always, the decision to write a monograph comes much later during the tenure of research in the relevant area. During the research period, quite a few papers have been published in various journals and conferences, following which these ideas have been extended further in the monograph. The publications that overlap with the contents of the book are given in the references [87–92, 145, 146]. Needless to say, the copy rights of various figures reproduced from these earlier publications belong to the original publisher. It may also be mentioned that some of the visualization outputs are in color. For various cost implications, we refrain from presenting them in color in the hardcopy version of the monograph. However the readers may opt for the e-book version for illustrations in color. The book is addressed to a broad readership. We expect graduate students and researchers in this area to find this book very useful. Application engineers in the remote sensing area will also find this monograph helpful. We have attempted to make the book as self-contained as possible. However, familiarity of readers with basics of image processing and statistical parameter estimation will help in better appreciating the contents of this book.Introduction Current State of the Art Edge-Preserving Solution Band Selection Through Redundancy Elimination Bayesian Estimation Variational Solution Optimization-Based Fusion Band Selection: Revisited Performance Assessment of Fusion Techniques Results and Discussions Conclusions and Directions for Future Research
Чтобы скачать этот файл зарегистрируйтесь и/или войдите на сайт используя форму сверху.