SPATIAL-ORGANIZATION IN 2D SEGMENTED IMAGES - REPRESENTATION AND RECOGNITION OF PRIMITIVE SPATIAL RELATIONS

Object recognition and scene analysis tasks can be greatly enhanced wh en information about spatial organization in an image is available. Mo reover, for recognition of complex objects a suitable representation o f spatial relations between objects' components taking into account sh ape, size, orientation, etc., is required. This cannot be accomplished by reducing a region to one or a few representative points; instead t he region as a whole must be treated. This paper presents a fuzzy logi c approach to the representation and recognition of spatial relations between regions in a 2D image. The main source of information on spati al relations is the geometry of the regions in question and we argue t hat this is complex enough to cause ambiguity in spatial relations, an d hence to warrant a fuzzy logic approach. The basic idea is to calcul ate the angles between the line connecting two points (one in each reg ion) and the horizontal line, to construct a histogram of these angles , and then upon an interpretation of the histogram as a fuzzy set to m atch it with the fuzzy sets representing a vocabulary of spatial relat ions. Other expressions of the spatial information which may be contex t dependent can be easily obtained by adding context knowledge. Severa l examples are used to illustrate our approach.