Abstract
Abstract
The article shows an application of satellite sensing data method in geoenvironmental monitoring of water surface. It is expected to apply combination of LBP and neural network approaches for detection and identification objects of natural and anthropogenic origin. The applying of satellite images, the implementation and operation of the filtration method and satellite sensing data assimilation in real or near-real time are considered to detect the blooming areas and their coordinates. The research demonstrates the need and possibility to apply neural approach and the theory of deep learning for solving the tasks. The results of computer experiments are presented basing on the images from satellites Resurs-P, WorldView and Landsat over the Azov sea area.
Subject
General Physics and Astronomy
Reference26 articles.
1. Conjugate equations and iterative algorithms in the problems of variational data assimilation;Marchuk;Proceedings of the Institute of Mathematics and Mechanics, Ural Branch of the Russian Academy of Sciences,2011
2. The Black sea as a simulation ocean model;Marchuk;Russian Journal of Numerical Analysis and Mathematical Modelling,2012
3. Data- computing technologies: a new stage in the development of operational oceanography;Marchuk;Izvestiya. Atmospheric and Oceanic Physics,2013
4. 4. Sensitivity of functionals of variational data assimilation problems;Shutyaev;Doklady Mathematics,2019
5. Methods for observation data assimilation in problems of physics of atmosphere and ocean;Shutyaev;Izvestiya Atmospheric and Oceanic Physics,2019