Open AccessProblematic Cellular Automata Segmentation and Clusterization of a Region’s Geoinformation Space
Author(s) -
Sarkis Artavazdovich Anesyants,
Alexander Belyaev,
С. О. Крамаров,
Владимир Храмов,
Daniil Chebotkov
Publication year - 2021
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/937/4/042074
Subject(s) - computer science , cellular automaton , cluster analysis , theoretical computer science , geographic information system , linear subspace , segmentation , data mining , fuzzy logic , information retrieval , artificial intelligence , mathematics , geography , pure mathematics , remote sensing
We consider the problems of clustering and segmentation for objects in the geoinformation space using the cellular automata theory, both classical and non-orthogonal ones. We clarify the terminology associated with the use of hybrid software and hardware for processing information coming from sources of different physical nature. This research is based on the geometric clusterization methods of multidimensional real or virtual spaces. As illustrative examples we consider two and three-dimensional variants, which, from our point of view, does not reduce the results’ significance in relation to the space of a greater dimension. Based on the formation conditions of the geoinformation space model as a semantic system, the use of semantic interoperability of its properties and corresponding subspaces is justified. It is shown that the unified geographic information space (UGIS) can be the data source for the formation procedures of various problem-oriented clusters used to manage socio-economic objects. As a variant of the UGIS formed subspaces this study uses a digital plan-diagram that has proven its effectiveness during previous works on the analysis of territories during their space monitoring. We also pay attention to the use of fuzzy methods and models in the processing of fuzzy source data and the clusters formation. Specific examples of clustering and segmentation using classical and non-orthogonal cellular automata are given.