Speaker
Description
The use of lake reeds for green energy is becoming an important component. It is important to determine the extent of reed areas, which in the future provides a forecast of the potential biomass volume and calculation of the carbon footprint. There are currently no significant results of research on the availability of reeds in Latvia. The main goal of this study is to develop a method for modeling the distribution of lake reeds to predict their availability in the future. The object of the study is satellite data of lakes, using the QGIS tool for modeling. The classification stage is very important, which allows distinguishing reeds from other surrounding objects. For the purpose of classification, several semi-automatic plugin algorithms were used: Minimum Distance, Random Forest, etc. The results of these algorithms were experimentally compared, and the most optimal one was selected.
Keywords — reeds, classification algorithms, modeling, satellite data, QGIS