An experiment on automated statistical recognition of clouds (Q1571284)

In this paper the authors present some results on the recognition of multispectral satellite data by means of a statistical automated classification algorithm. The algorithm is based on the approximation of an unknown probability density function for a given set of observations by a mixture of multi-dimensional normal distributions. For a given number of mixture components, optimal estimates for unknown parameters are found by the Day-Shlezinger algorithm as one of the solutions of the set of likelihood equations that maximize the likelihood functions.