Template matching techniques in computer vision: theory and practice

This book presents a unified approach to pattern recognition techniques for face recognition and interpretation proposed so far by the computer vision community. There is a particular emphasis on template matching and employing the idea of projection to match image patterns on a specific class of target objects (faces), in order to illustrate general object recognition approaches. The author starts with an introductory chapter, setting out the context for template matching techniques and the goals of computer vision in general. The following chapter looks at the link between human and computer perspectives, before examining the basics of digital image formation the theoretical basis for imageprocessing, including the physics of light, lenses, sensors, linear computation and similarity. The remainder of the text explores the issues behind building a face recognition system from the ground up. Brunelli uses a virtual vision approach throughout, to present, evaluate and compare algorithms using computer generated, photo-realistic images. Concluding with appendices on syntheticimagery in computer vision research, there is also a supplementary website containing sample algorithm implementation examples and image datasets so that the reader can perform practical tests and understand the qualitative and quantitative aspects of the material presented.

• ISBN: 978-0-470-51706-2
• Editorial: John Wiley & Sons
• Encuadernacion: Cartoné
• Páginas: 348
• Fecha Publicación: 20/03/2009
• Nº Volúmenes: 1
• Idioma: Inglés