Formation of an informative index for recognizing specified objects in hyperspectral data

The paper is about the development of an approach which able to create rules for distinguish-ing between specified objects hyperspectral data using a small number observations. Such would contribute methods and algorithms operational analysis data. These can be used preprocessing labeling. Implementation proposed are technology that harnesses both discriminative criteria general formulas spectral indexes. In implementing technology, index was defined with normalized difference formula. Informativeness estimated separability analysis. results show implemented algorithm recognize areas different vegetation. formed by similar Normalized Difference Vegetation Index (NDVI). generalization forming recognition features. It has been shown form informative indexes in tasks