Overenie možností vyhľadania zmien polohovej situácie v priestore letiska multitemporálnou analýzou rastrových a vektorových podkladov

Abstract

The thesis focuses on exploring the use of Deep Learning technology in combination with geodesy and GIS systems in the field of aviation for the purpose of updating digital data of selected airports. The main goal is to determine whether it is possible to create a Deep Learning model capable of accurately and correctly detecting selected airport structures. The theoretical part provides information necessary for understanding the fundamentals of the Deep Learning method and closely related concepts such as Machine Learning and Artificial Intelligence, fundamentals of aviation and airport data as well as information about the collaborating company NG Aviation. The practical part discusses the workflow, including a justified selection of airports and their brief descriptions, followed by the core of the thesis, which offers a detailed description of the entire model development process. In conclusion, the overall accuracy and reliability of the chosen method are evaluated, along with an assessment of its suitability for the given task.