Modern Algorithms for Image Processing (eBook)

Vladimir A. Kovalevsky holds a diploma in physics, a PhD in technical sciences, and a PhD in computer science. He has been a researcher, professor, and visiting professor at many esteemed universities worldwide, including the Central Institute of Cybernetics of the Academy of Sciences, University of Applied Sciences, and the Manukau Institute of Technology. Dr. Kovalevsky has been a plenary speaker at many conferences and his research interests include digital geometry, digital topology, computer vision, image processing, and pattern recognition. He has published four monographs and more than 180 journal and conference papers.

Part I: Image ProcessingChapter 1: Introduction to Image ProcessingChapter 2: Digital Image FundamentalsChapter 3: Image TransformsChapter 4: Image EnhancementChapter 5: Image RestorationChapter 6: Image CompressionPart II: Image AnalysisChapter 7: Image SegmentationChapter 8: Representation and DescriptionChapter 9: Recognition and InterpretationChapter 1 describes elements of image processing systems.Chapter 2 describes the structure of the human eye, basic ideas of sampling and quantization and relationships between pixels. It describes among other the simplest algorithm for labeling connected components in binary images. In Chapter 8 the authors come back to connected components and they describe a method of labeling connected components by means of morphology. But they do not describe a method labeling all components of an image in one go as my "Root Algorithm" described in Section 9.3.

Chapter 3 describes the Fourier transform and some other transforms.

Chapter 4 describes enhancement method in spatial and in frequency domains. The filters, also the median filter, are described very shortly, without mentioning details. This chapter contains a detailed presentation of the fundamentals of color images. The processing of color images is described very shortly with saying that an RGB image should be converted to a HSI image and then its intensity component can be processed with methods developed for gray-level images. My book contains on many places remarks to the difference between processing color and gray-level images.

Chapter 5 describes a very interesting problem of the restoration of degraded images. The authors understand under degradation mostly the blurring of an image (without saying this, which makes the reading difficult). Interesting results of restoring blurred images by means of Wiener filter are presented.

Chapter 6 is devoted to image compression. A statistically based theory of the compression is presented, and different methods of compression are described. However, among the described methods there is no one whose idea is similar to the idea of my method of compression (Section 8 of my book).

Chapter 7 describes different methods of image segmentation but not the segmentation by quantization of colors as described in my Section 9. In this chapter they describe also some primitive methods of edge detection but they do not mention the modern Canny edge detection in spite that the work of Canny was published 7 years before the book was printed.

Chapter 8 covers a primitive procedure for the polygonal approximation of boundaries. This procedure has the drawback that it is impossible to control the precision of the approximation and that it will not correctly work in the case of boundaries having deep indentations. My approximation described in Section 11 is free from these disadvantages.

Chapter 9 describes different methods of recognizing geometric shapes, but no precise methods, such as my described in Sections 12 and 13.