An Intellectual Approach to Segmentation of the Satellite Images

Workshop on Materials and Engineering in Aeronautics

材料科学;航空航天工程

Demidova, L.A.^1^2 ; Tishkin, R.V.^3

Ryazan State Radio Engineering University, Ryazan, Russia^1

Moscow Technological Institute, Moscow, Russia^2

Limited Liability Company Areal-98, Moscow, Russia^3

关键词: Remote sensing satellites;    Satellite images;    SVM classifiers;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/476/1/012008/pdf
DOI  :  10.1088/1757-899X/476/1/012008

来源: IOP

【 摘 要 】

The applicability of the clustering algorithms under uncertainty conditions for the image segmentation from the remote sensing satellites has been shown. Approaches to expanding the characteristics' space to improve the quality of the image segmentation have been considered. An intellectual approach to improving the quality of the image segmentation, which consists in combining the results of the image segmentation from the different clustering algorithms under conditions of uncertainty through their use in the development of the SVM classifiers and their ensembles, has been formulated. The results of solving the problem of the image segmentation based on the clustering algorithms under uncertainty conditions have been presented.