This book is the best textbook on image processing for senior/graduate students majoring in engineering or computer science. Although a knowledge of calculus and linear algebra is presumed, it is a very accessible textbook. Chapters one and two consist of very basic background information. The concepts of linearity, pixel distance measures, spatial versus gray scale resolution, and zooming and shrinking are explained. Chapter 3 is about image inhancement in the spatial domain, and includes discussions on contrast enhancement, histogram processing and equalization, and histogram matching. The idea of filtering images via an NxN kernel mask is also introduced. Chapter 4 is about filtering in the frequency domain. The 2D Fourier transform is introduced and it is explained how filtering can take place using this transform. Chapter five discusses image restoration. This includes Weiner filtering and minimum mean square error filtering. Chapter six discusses color image processing. This chapter discusses the various color spaces - RGB, CMYK, HSI, and how the transforms mentioned up to this point in the book can be performed in color. Chapter 7 is about wavelets and multiresolution processing. This chapter is a good solid presentation of wavelets and their usage in image processing. I would suggest that anyone interested in this subject start here before they read another book, since the presentation is clearer here than in books dedicated to the subject. Chapter 8 is about image compression. Basics of information theory are discussed, and lossy as well as lossless methods of compression are discussed. As a good follow-on to the previous chapter, the role of wavelets in compression is discussed. Chapter 9 discusses morphological image processing, which is that field of image processing that relies on the systematic "fattening" and "thinning" of edges to enhance images. Chapters 10,11, and 12 are a sort of introduction to computer vision topics. Chapter 10 discusses how to segment an image. Chapter 11 is about image descriptors that quantify segmented portions of an image. Chapter 12 is about object recognition and even has a short section on statistical classifiers. This book is a joy to read, and will make the topic of image processing very clear. There are plenty of diagrams, formulas, and equations listed. There are no examples to speak of, but algorithms are clearly specified so that I don't think that the book suffers because of the lack of examples. All engineering textbooks should be this well written. I particularly recommend this book as a reference for students and practitioners of robotics, video processing, and computer vision, since there are image processing considerations in all of these fields that this book will clarify.
