Customer Reviews

Image Processing and Data Analysis: The Multiscale Approach

5 star:		(2)
4 star:		(0)
3 star:		(0)
2 star:		(1)
1 star:		(0)

 

 

From my persepective I consider this book a bargain. The price is right and I certainly got my nickels worth reading it.

I am heavily into 2-D wavelet analysis of remote sensing imagery. My initial,naive foray into this topic was a very frustrating experience. Most wavelet books do not cover multidimensional applications well or at all. In fact, most wavelet books are inordinately concerned with the mathematical nuances of the subject and pay little attention to the mechanics of implementing the theory. So all wavelet books read pretty much the same and there is usually nothing new or insightful from one text to another.

Starck et al is the exception. It has the only clear exposition of 2-D non-separable and a trous wavelet methods that I have found anywhere. The book has been a time saver for me and not a time waster, and well worth the investment.

This book appeared very easy to read and is an excellent resource of multiscale techniques. The practical applications that are discussed in this book combined with the theory, give a good overview.

I purchased this book because I am interested in applying wavelet to financial time series. The authors have written a paper in using wavelets and noise analysis to filter these time series, so I thought that this book had promise. And in fact chapter 2 deals with noise modeling and noise filtering.

The only problem is that I found this material impossible to follow. Despite the claims that the book is aimed at applications, there are few real examples. In the case of noise modeling and noise filtering there are none. It was impossible for me to tell whether the techniques they seem to suggest work or how they can be applied.

I've become fairly expert on Haar wavelets. Even so I can read their explaination of the Haar wavelet transform only with difficulty.

The authors give the impression that they are academics writing for academics (at the post graduate level). The book provides equations as quick short hand for ideas and the reader is expected to fill in the blanks. Perhaps a post graduate background in math is needed. However I would not say that the publishers description properly reflects the contents of this book: "[the book] develops the reader's undersanding of each technique and then shows with practical examples how they can be applied to improve the skills of [...] electrical engineers." Real world applications are mentioned, but never fully explained.

All in all I found this book an expensive disappointment.