Segmentation and classification of X-ray diffraction patterns by inhomogeneous synchronized chaotic maps

A new approach to the classification and segmentation of X-ray two-dimensional patterns is described. It aims at clustering a huge dataset, e.g. collected at the synchrotron radiation facilities, by means of the synchronization of chaotic maps, whose evolution is driven by the experimental data. The method has revealed robust and precise and, furthermore, it does not require any additional input (e.g. number of clusters). It has been successfully applied to the analysis of X-ray diffraction data of biomaterials.