Introduction

Abstract

The term image understanding has been widely used since work on image understanding in the United States started in 1975. It refers to knowledge-based interpretation of visual scenes by computers and consequently denotes an interdisciplinary research area including signal processing, statistical and syntactic pattern recognition, artificial intelligence, psychology, and even neurophysiology. In the early literature, it was called scene analysis. Computer vision is also widely used to refer to a similar research area; but while computer vision emphasizes computational aspects of visual information processing, such as measurement of three-dimensional shape information by visual sensors, image understanding stresses knowledge representation and reasoning methods for scene interpretation. Although understanding time-varying scenes is a very important topic in image understanding and computer vision, in this book we confine ourselves to image understanding of static scenes.