Обзор различных методов отбора наиболее информативных спектральных диапазонов

Multispectral data provide capability for discrimination between many land cover types. But the great number of spectral bands is problematic for statistical classification techniques. In this article we considered some methods for selection of the most effective and most informative spectral bands. In this work we considered Band Selection method for Hyperspectral Data Analysis and Automatic Clusterization. It was also proposed a new approach to select spectral bands, which is based on the concept of criterion function of information capability of spectral bands. In this article we also considered and analysed some examples with use of criterion function of information capability.