A processing model for two-dimensional images based on the object-oriented paradigm

The object-oriented paradigm is considered a natural and effective approach for image processing and analysis aims. Nevertheless, the use of object orientation within this complex realm has not undergone a rapid growth, as it happened in other fields where the paradigm has been introduced. We argue that one of the major motivations stems from the difficulty of conceiving a comprehensive framework for coping with the different abstraction levels and the several operations that an automated vision task requests. With the aim of overcoming such major drawback, in this paper we present a new model for the processing and the analysis of two-dimensional images based on the object-oriented paradigm. The model states an object-founded representation of images, it provides the possibility of handling heterogeneous data structures within the same image, and it designs image processors on the basis of both the protocol of interaction with images, and their processing behavior. Examples are given and results are discussed with respect to related work in the literature.