Využitie umelej inteligencie na automatizované rozpoznávanie objektov z rôznych typov rastrových dát

Abstract

The main objective of this thesis is to analyse and implement selected machine learning approaches, in particular deep neural networks, for the identification and classification of objects in image data. The thesis focuses on the basic principles of image data processing, the design and implementation of a model, its training, and the evaluation of its accuracy. The result of the work is the verification of the applicability of artificial intelligence for automated object recognition in geodetic and geoinformation applications. The thesis also highlights the potential of these methods for more efficient processing of large volumes of spatial data and their use in practice, for example in updating geographic databases or monitoring changes in the landscape.