N. N. Aprausheva ,   I. A. Gorlach ,   A. A. Zhelnin ,   S. V. Sorokin Computational Mathematics and Mathematical Physics , Vol. 38, No. 10 , 10.1998 , p. 1715-1719, language: английский  Аннотация Results of the recognition of multi-spectral satellite data by means of a statistical automated classification algorithm (SACA) are presented. 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 function. Optimal number of classes is determined by the step-by-step checking of two complex statistical hypotheses. The classification of a given set of observations is performed by applying the bayesian rule. To reduce the computational cost of the SACA, a preliminary analysis of the structure of the set under investigation is carried out, which provides rough estimates for the number of classes and their basic characteristics. Results of automatic classification of the main types of clouds and underlying surface are described.   Полный текст     в формате pdf