Information technologies of simulation of beam structures using multispectral spacecraft materials

Abstract

The work is devoted to solving the scientific and practical task of modeling beam structures based on the materials of multispectral space images based on aerospace and contact measurements. The effectiveness of the use of space survey materials for the study of soil cover largely depends on the time of the survey. The linear forms of erosion (waterholes, ravines) are deciphered on space photographs, which are displayed only on images with a spatial resolution of 1-2 meters in the form of narrow, clearly delineated contours that have a jagged shape.
The developed clustering algorithm makes it possible to more clearly identify beam structures on satellite images in combination with data from digital terrain models (DRM). A number of experiments were conducted on some set of remote sensing data of Boryspil and c. Panchevo, Novoukrainsky District, Kirovohrad Region. The experiments consisted in determining the spectral and other features of the beam structures (for example, the shape, soil types, etc.), which are located in different places of the territory that was investigated.
On images with a spatial resolution of more than 10 meters, ravines are usually not displayed, but networks of beams with elongated wavy tree-like shapes are clearly visible. In the course of the study, data from multispectral imaging from the Sentinel-2 satellite (MSI scanner) and DEM (Digital Elevation Model, DEM) data obtained through the SRTM 3 archive were used.
The research results showed that automating the process of highlighting beam structures on multispectral images with a spatial resolution of 30 m is very difficult. First of all, this can be explained by the fact that the beam consists of vegetation that is also present outside the beam. In addition, the number of channels of the ETM+ scanner and its spatial ability are not enough for a clearer separation of the "beam" class.