Classification of Norway spruce drying states on the basis of reflection spectra

The article is devoted to the development of a method for classifying reflection spectra spruce needles different health categories and assessing quality classification. Such spruces have been chosen that include initial stages drying out, which essential early detection disease foci, but makes it difficult classify according visual criteria by means remote sensing Earth. An algorithm untrained classification visualisation spectral data based on correlation cluster analysis proposed. obtained under laboratory conditions were studied results interpreted using developed software. various combinations parameters within proposed algorithm, as well individual components with known methods, made possible determine most effective combination methods (projection into space principal components, elimination influence first component spectra, Ward linking metric standardised Euclidean calculating distance) disease. Its use increase F-score indicator 2nd category (the important task detecting in stages) up 70.59 %.