PE&RS December 2004
VOLUME 70, NUMBER 12
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
JOURNAL OF THE AMERICAN SOCIETY FOR PHOTOGRAMMETRY AND REMOTE SENSING
Book Review
Foundations of Image Science
Barrett, H.H. and Myers K.J. John Wiley & Sons, Inc: Hoboken,
New Jersey
Xli and 1540 pp., diagrams, mathematical appendices, bibliography,
index.
$140.00. Hardcover. (ISBN 0-471-15300-1).
Reviewed by: Geoff Dougherty, Professor, Physics and Medical Imaging,
California
State University Channel Islands.
This book is intended to become a standard text in the field of imaging
science. Its target audience is primarily graduate students and
practitioners working with imaging systems, and well-motivated
undergraduate students. There are a number of themes within the
book, the major one being the need for objective assessment of
image quality. All the theoretical tools for handling these themes
are
included: the principles, the mathematics and the statistics. A thorough
treatment of the continuous-to-discrete model of digital imaging
is presented throughout, and numerous examples are used and
applications developed.
The book is grandiose in its vision. Its purpose is to provide an
underlying theoretical foundation, based on both theory and experiment,
to tie together all the sub-disciplines of imaging science. The
central thesis is that image quality needs to be defined objectively,
measured experimentally and optimized iteratively according to fundamental
principles rather than ad hoc notions. Its unifying theme is
the treatment of the imaging process as a mapping between object
space and image space, with the continuous-to-discrete transform
being the most common. This treatment is applied comprehensively
to a wide variety of imaging systems, from optical imaging to planar
imaging in radiography and nuclear medicine to tomographic reconstruction
and more. Along the way, the authors present the detailed“
tools of the trade” in terms of mathematics and methodology,
and
cover multifarious themes associated with the big picture of imaging
science as an integrated whole. The goal is not only to understand
all
the aspects of the process so as to use them objectively in assessing
an imaging task, but to be able to iteratively optimize the process.
The accuracy of the material cannot be faulted, and the book approaches
its material in a straightforward no-nonsense style. Its arguments
are convincing and well-supported by examples. The authors
are eminent practitioners in the field, and bring together many
decades of practical experience, which enlightens and illuminates
the text. Their attention not only to detail within each part of
the
argument, but also in their explanation as to how these parts contribute
to the big picture of image formation, representation and
assessment, is to be applauded. This is not a text for the faint-hearted.
The mathematics is rigorous and pervasive and can be overbearing
at
times, but not to the point where the message becomes obscured.
For a textbook on imaging, it might have been preferable to get past
some of the initial chapters on the mathematical tools more quickly
and get into the core concepts earlier: for a reference book, this
is not as big a problem. Since it is about imaging, I would have
liked
more images to illustrate the concepts, particularly in the discussion
of noise and image quality, and to make it more accessible.
Foundations of Image Science is certainly an impressive tome,
comprehensive and inclusive in its sweep of imaging systems and
breath-taking in its proportions. The authors approach the subject
with a missionary zeal to collect the sometimes disparate approaches
to imaging and forge them into an integrated whole. It is a scholarly
work, put together with great attention to detail. I suspect that
it
will become a reference work in the field, rather than a standard
textbook. The sheer size, not
to mention weight, mitigates
against the latter eventuality.
Indeed, I would have preferred
to have seen it split
into a three-volume text,
which would have made it
less daunting and more portable.
Indeed, the authors
have it in them to expand the
material into even more volumes,
as they allude to in the
Epilog. I would have liked to
have seen them tackle ultrasound
and magnetic resonance
imaging. Incidentally,
I found that the binding is insufficiently
strong for a book
of this magnitude: my copy
is already splitting apart at
the seams.
In conclusion, the book
certainly fills a void in the
available literature with its
comprehensive and unified
treatment of imaging systems
and science and is destined
to become a reference
work in this field. I highly recommend
this book.
|