Processing of Multispectral Survey Materials based on Machine Learning Methods for Forest Management

The paper consider the machine learning methods application for processing multispectral space imagery materials. The forest management is carried out on the basis of a forest vegetation regular analysis. Identification of the forest vegetation type and its condition assessment are carried out by machine learning methods based on spectral-brightness features. The training and test samples choice is provided. We presents comparative analysis of machine learning methods for solving forest areas classification and clustering problems for further justified control action choice.