In a previous work, we defined a new approach to evaluating directional relations between objects (such as ``is to the right of'', ``is above''...). The approach is based on the c omputation of a histogram of forces. We show in the present paper that the notion of the histogram of forces ca n also be useful in pattern recognition. The aim of this study is to classify orbits and sinuses. Using force histograms instead of geometrical criteria (like compactness or degree of ellipticity) allows more accurate classification to be produced. Moreover, the proposed approach is low computing ti me.