Qualitative Models of Shape, Size, Orientation and Distance Applied to the Description of Images Containing 2D Objects

In this paper, we present an approach to qualitative description of images containing non-overlapping 2D objects. This approach consists of extracting the properties of the main points of the objects in an image and applying qualitative models to them. Qualitative models of shape and size describe visual features of the objects in the image, while qualitative models of orientation and distance describe spatial features of those objects. Finally, a string which describes the image in qualitative terms is obtained. This string can be used to compute the relationships between the objects in the image and also to compare images by obtaining a degree of similarity between them