Image Models

Narendra Ahuja and B. Schachter

Abstract

Models of spatial variation used in the computer processing of pictorial information are surveyed. Models of images depicting homogeneous textures are reviewed under the categories of pixel-based and region-based models. Pixel-based models are further divided into one-dimensional time series models, random field models, and syntactic models. The random field models incorporate either global or local properties of an image. To put the role of the two low-level models in perspective, several high-level models making use of semantic information are also described.