Search academic texts

Information × Registration Number 0218U003964, 0118U001655 , R & D reports Title Deep Leaming models and methods for geospatial analysis in agriculture domain popup.stage_title Head Andrii Shelestov, Registration Date 04-10-2018 Organization National Technscal University of Ukraine "Kiev Polytechnic Institute". popup.description2 The object of research — high resolution satellite imagery methods of their processing. The main goal of the work — this project is dedicated to the development of modern scientific models and methods of deep learning based on artificial intelligence, allowing to process big geospatial data with different nature (terrestrial, satellite from different space systems) and solving the actual tasks of classification of the Earth's surface and agricultural crops. The verification results of the developed methods will be correspondent to the modern requirements for the processing of geospatial information and the principles adopted and accepted by the JRC (Joint Research Center) for the geospatial data usage in the interests of state and local authorities. Method of research — machine learning, geospatial analysis, object-oriented design, structural functional analysis, statistical methods, neural networks. Within the project were proposed approaches for collecting a training samples and further digitization based on satellite data. They are based on the data analysis from spectral channels of satellite images and collected ground surveys during the formation of the sample. It’s proposed to use optical and radar satellite data for the crop’s classification. Product Description popup.authors А.М. Лавренюк Г.О. Яйлимова М.С. Лавренюк О.О. Напреєнко popup.nrat_date 2020-04-02 Close