A Nonparametric Pattern Recognition Algorithm in the Problems of Analyzing the Data of Remote Sensing over Anthropogenic Territories
摘要
Abstract
Some results from the application of a new method of verifying the hypothesis about the independence of random variables in the analysis of remote sensing data for anthropogenic territories are reported. The method is based on a nonparametric pattern recognition algorithm corresponding to the maximum likelihood criterion. Linear and nonlinear dependences between the spectral features characterizing anthropogenic territories are determined. The results of recognizing the types of anthropogenic territories by the spectral data of remote sensing are considered.