- ホーム
- > 洋書
- > 英文書
- > Science / Mathematics
Full Description
Visual Communication: An Information Theory Approach presents an entirely new look at the assessment and optimization of visual communication channels, such as are employed for telephotography and television. The electro-optical design of image gathering and display devices, and the digital processing for image coding and restoration, have remained independent disciplines which follow distinctly separate traditions; yet the performance of visual communication channels cannot be optimized just by cascading image-gathering devices, image-coding processors, and image-restoration algorithms as the three obligatory, but independent, elements of a modern system. Instead, to produce 'the best possible picture at the lowest data rate', it is necessary to jointly optimize image gathering, coding, and restoration. Although the mathematical development in Visual Communication: An Information Theory Approach is firmly rooted in familiar concepts of communication theory, it leads to formulations that are significantly different from those that are found in the traditional literature on either rate distortion theory or digital image processing.
For example, the Wiener filter, which is perhaps the most common image restoration algorithm in the traditional digital image processing literature, fails to fully account for the constraints of image gathering and display. As demonstrated in the book, digitally restored images improve in sharpness and clarity when these constraints are properly accounted for. Visual Communication: An Information Theory Approach is unique in its extension of modern communication theory to the end-to-end assessment of visual communication. from scene to observer. As such, it ties together the traditional textbook literature on electro-optical design and digital image processing. This book serves as an invaluable reference for image processing and electro-optical system design professionals and may be used as a text for advanced courses on the subject.
