The aim of this paper is the design of a system that can learn any
individual user’s perception of four subjective and key spatial
relationships between image objects: “to the right of,” “above,” “to
the left of ” and “below.” The proposed approach is based on the
utilization of artificial Neural Networks (NNs) and the modeling of
relative positions between 2D objects through histograms of forces. The
NNs are fed by features extracted from the histograms and are trained
to numerically assess the four relationships according to the target
perception.
|