Review by
Simon Gibson, University of Manchester, UK.
Computer Graphics Forum
(The International journal of the Eurographics association),
vol.16, n.5 - Dec/1997
pp. 307-308
Although many previous books have been written on digital image processing, "Image Processing for Computer Graphics", as it title says, presents material that is important to people working in the fields of visualisation and computer graphics. The book lies somewhere between a tradicional image processing text, covering theoretical basics such as signal processing and image sampling and reconstruction, and a computer graphics book, with sections on morphing, digital compositing and colour quantisation, and as such fails to cover either area entirely satisfactorily.
The emphasis in each chapter is on mathematical concepts, and the book aims to provide a conceptual understanding of the image processing operations most commonly used by computer graphics practitioners. The authors do admit that some mathematical rigour has been sacrificed in places, with the aim of providing more intuitive descriptions, although no pseudo-code of any kind is given throughout the book.
Starting with an introduction to both image processing and computer graphics, the book then delivers the mandatory chapters on signal theory, operations on images, and sampling and reconstruction. Also included are sections on the fundamentals of colour theory and its various methods of representation, on computers. Following are chapters on image representation, compositing, dithering and half-toning, as well as brief discussions of warping and morphing.
For the student interested in the more theoretical aspects of digital images, this book provides a useful summary of the relationship between image processing and computer graphics. For anybody who has any fairly general texts on eigther topic on their bookshelves already, little new information is given, and the lack of pseudo-code means that the beginner may struggle to put much of the information to any practical use.