SEMI-AUTOMATIC ALGORITHM FOR CONSTRUCTING THE CONTOUR OF AREA OBJECTS ON MULTISPECTRAL SATELLITE IMAGE

Abstract

This article formalizes the problem of semi-automatic construction of the contour of area objects from satellite multispectral images and presents a solution algorithm using PCA and Dijkstra's algorithm. The emphasis of the article is on optimizing PCA using annual matrices for images of the same class, as well as optimizing the Dijkstra algorithm using dynamic programming methods. The contour is considered as the boundary of an object, which can be used for its segmentation and classification. The semi-automatic contour accepts reference points specified by the operator. The formalization of the algorithm is completed.