Abstract
Support of the economic growth of Ukrainian agriculture requires the development and effective use of innovative technologies. In particular, satellite analysis makes it possible to monitor the state of agricultural land by monitoring their qualitative and quantitative indicators of natural and climatic characteristics. Satellite monitoring of agricultural land use in Ukraine has been developed within the World Bank program “Supporting Transparent Land Governance in Ukraine” in collaboration with EOS Data Analytics and Space Research Institute National Academy of Sciences of Ukraine and State Space Agency of Ukraine. Based on the developed technology, classification maps of the land cover were built based on three data sets: ground data along roads, farmers' data, and satellite data (time series of Sentinel-2 optical data and Sentinel-1 radar data). To create classification maps, the Random Forest algorithm was used, which is implemented on the Google Earth Engine cloud platform. An accuracy assessment was carried out, and crop compared areas throughout Ukraine were obtained. According to the results of the experiment, a comparison of the classification obtained from two separate training data sets (ground data collected along roads and data of farmers) is given. As a result, a validated crop map was obtained. The map is presented on the official web-portal of the State Geocadastre of Ukraine. The main results of the analysis of the agricultural lands of Ukraine, as well as, the results of comparisons with statistical data, are presented.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
