CASPIAN JOURNAL
MANAGEMENT AND HIGH TECHNOLOGIES
Algorithm of semantic classification of spatial objects and scenes with use of topological features and the random forest qualifier on multiscale maps
 |
Read | Andrianov D. E., Eremeev S. V., Kuptsov K. V. Algorithm of semantic classification of spatial objects and scenes with use of topological features and the random forest qualifier on multiscale maps // Caspian journal : management and high technologies. — 2018. — №4. — pp. 8-18. |
/8-18.jpg)
Andrianov D. E. - Doct. Sci. (Engineering), Head of the Department, Murom Institute (branch) Federal State Budgetary Educational Institution of Higher Education"Vladimir State University named after Alexader Grigoryevich and Nickolay Grigoryevich Stoletovs", 23 Orlovskaya St., Murom, Vladimir region, 602264, Russian Federation, andrianovde@inbox.ru
Eremeev S. V. - Cand. Sci. (Engineering), Associate Professor, Murom Institute (branch) Federal State Budgetary Educational Institution of Higher Education"Vladimir State University named after Alexader Grigoryevich and Nickolay Grigoryevich Stoletovs", 23 Orlovskaya St., Murom, Vladimir region, 602264, Russian Federation, sv-eremeev@yandex.ru
Kuptsov K. V. - post-graduate student, Vladimir State University named after Alexander and Nikolay Stoletovs, 87 Gorky St., Vladimir, 600000, Russian Federation, kirill-kuСЂtsРѕv@rР°mblРµr.ru
The problem of classification of spatial distributed objects and groups of objects on multiscale raster maps is considered in article. The theory based on methods of a persistent homology and computer topology is the cornerstone of the solution of the considered tasks. The existing methods of classification of spatial objects are considered. The mathematical description of an algorithm of classification on the basis of topological features, simplicial complexes, barcodes and delta codes is provided. New operations over barcodes and delta codes are developed. The algorithm of classification of the spatial distributed objects is presented. It allows to define the general structure of spatial objects on multiscale maps without loss of key spatial information. Results of a research of work of an algorithm for multiscale cards are presented
Key words: классификация групп объектов, классификация пространственных объектов, обработка изображений, персистентная гомология, пространственная сцена, топологический анализ данных, топология, classification of groups of objects, classification of spatial objects,