NONPARAMETRIC PATTERN RECOGNITION ALGORITHM IN TASKS OF ANALYZING REMOTE SENSING DATA OF ANTHROPOGENIC TERRITORIES
A. V. Lapko1,2, V. A. Lapko1,2, A. V. Sharueva2
1Institute of Computational Modelling, Siberian Branch, Russian Academy of Sciences, Krasnoyarsk, Russia
2Reshetnev Siberian State University of Science and Technology, Krasnoyarsk, Russia
Keywords: testing the hypothesis of the independence of random variables, pattern recognition, nonparametric estimation of probability density, anthropogenic territories, remote sensing
Abstract
The results of the application of a new methodology for testing the hypothesis of the independence of random variables in the analysis of remote sensing data of anthropogenic territories are presented. The basis of the technique is a nonparametric pattern recognition algorithm corresponding to the maximum likelihood criterion. Linear and nonlinear dependences between the spectral features that characterize anthropogenic territories are determined. The results of the assessment of anthropogenic territories based on spectral remote sensing data are considered.
|
