Comparison of different approaches to analytical images classification

Classification is a powerful tool for the extraction of chemical informatio n from analytical images, especially for the distinction of chemical phases and the identification of transient phases. There are various approaches t o classification and numerous algorithms exist. This paper aims to examine three different types of classification algorithms with respect to their su itability for analytical image classification: the Minimum Distance Algorit hm as a statistical method, the Kohonen Net Classification as a Neural Netw ork approach and the Fuzzy c-means Clustering Algorithm as an example for a Fuzzy Logic approach. This examination is performed with test images for t he quantitative aspect and with real analytical images for the qualitative aspect.