Directional relative position between objects in image processing: a comparison between fuzzy approaches

Abstract

The importance of describing relationships between objects has been highlighted in works in very different areas, including image understanding. Among these relationships, directional relative position relations are important since they provide an important information about the spatial arrangement of objects in the scene. Such concepts are rather ambiguous, they defy precise definitions, but human beings have a rather intuitive and common way of understanding and interpreting them. Therefore in this context, fuzzy methods are appropriate to provide consistent definitions that integrate both quantitative and qualitative knowledge, thus providing a computational representation and interpretation of imprecise spatial relations, expressed in a linguistic way, and including quantitative knowledge. Several fuzzy approaches have been developed in the literature, and the aim of this paper is to review and compare them according to their properties and according to the types of questions they seek to answer.