Литинский П.Ю.
Классификация сканерных снимков методом моделирования спектрального пространства
// Труды КарНЦ РАН. No 5. Сер. Математическое моделирование и информационные технологии, вып. 2. 2011. C. 45-54
Litinsky P.Yu. Multispectral imagery classification method based on spectral space modeling // Transactions of Karelian Research Centre of Russian Academy of Science. No 5. Mathematical Modeling and Information Technologies. 2011. Pp. 45-54
Keywords: remote sensing, multispectral imagery classification, geoinformation modeling.
A method for satellite imagery (Landsat TM/ETM+, IRS, etc.) classification based on spectral space modeling is described. The 3D model is built in xyz-axes, where x and y are the first two principal components of the image matrix in logarithmic form (bands G, R, NIR, SWIR2), and z is the moisture stress index MSI (SWIR1/NIR). Ambiguous spectral classes are decomposed using the geomorphometric model and time series imagery. The method is optimal for taiga ecosystem structure and dynamics modeling.
