In this talk, we present an approach using Inductive Logic Programming (ILP) to automatically learn non trivial descriptions of symbols, based on a formal representation.

The overall goal of our approach is to express graphic symbols by a number of primitives that may be of any complexity (i.e. not necessarily just lines or points) and connecting relationships that can be deduced from straightforward state-of-the art image treatment and analysis tools. This representation is then used as an input to an ILP solver, in order to deduce non obvious characteristics that may lead to a more semantic related recognition process.

Since one of the major difficulties in classification is the assessment of classification quality, the presentation will contunue with a way of visually reconstructing the learnt classes in order to give a feedback to the user. The system translates positionning constraints into a linear programming problem, which can be solved with standard state-of-the-art approaches. The result can then be used for generating a synthetic image, representative of the class.