An interdisciplinary introduction to image processing : pixels, numbers, and programs / Steven L. Tanimoto.

Includes bibliographical references and index.

Getting started -- Brightness and contrast -- Controlling color -- Geometric transformations -- Geometric distortions -- Synthesizing images -- Stereograms -- Images within images -- Filtering -- Introducing python -- Basics of python -- Control structures : Conditionals and repetition -- Data structures -- Creating functions -- Programming techniques -- Image analysis -- Pattern recognition -- Computational photography -- Selected applications -- Calculator formulas -- Quick reference-PixelMath formulas -- Python glossary -- Troubleshooting -- Graphical user interfaces.

This book explores image processing from several perspectives: the creative, the theoretical (mainly mathematical), and the programmatical. It explains the basic principles of image processing, drawing on key concepts and techniques from mathematics, psychology of perception, computer science, and art, and introduces computer programming as a way to get more control over image processing operations. It does so without requiring college-level mathematics or prior programming experience. The content is supported by PixelMath, a freely available software program that helps the reader understand images as both visual and mathematical objects. The first part of the book covers such topics as digital image representation, sampling, brightness and contrast, color models, geometric transformations, synthesizing images, stereograms, photomosaics, and fractals. The second part of the book introduces computer programming using an open-source version of the easy-to-learn Python language. It covers the basics of image analysis and pattern recognition, including edge detection, convolution, thresholding, contour representation, and K-nearest-neighbor classification. A chapter on computational photography explores such subjects as high-dynamic-range imaging, autofocusing, and methods for automatically inpainting to fill gaps or remove unwanted objects in a scene. Applications described include the design and implementation of an image-based game. The PixelMath software provides a "transparent" view of digital images by allowing the user to view the RGB values of pixels by zooming in on an image. PixelMath provides three interfaces: the pixel calculator; the formula page, and advanced extension of the calculator; and the Python window.