Shape recognition combining mathematical morphology and neural networks

The purpose of this paper is to show that some adapted preprocessing can help neural networks in classifying binary planar shapes. After recalling some basic notions of mathematical morphology and some shape numbers, a morphological preprocessing transforming a binary planar shape into a curve which is translation, rotation, and scale invariant, namely, the spectral function, is presented. These extracted shape features are discriminated by statistical and neural methods. Finally, an application on real data is described.