Detekce zemědělských plodin s využitím dat Sentinel-2

Abstract

The bachelor thesis focuses on the use of Sentinel-2 satellite data for the detection of agricultural crops. The thesis is divided into theoretical and practical parts. The theoretical part describes methods of data collection through remote sensing of the Earth and the types of measurements used in this process. It further elaborates on the use of the NDVI index for the analysis and monitoring of agricultural crops and justifies the selection of areas for investigation. The thesis also presents classification methods, whose effectiveness was assessed using the confusion matrix and other metrics to verify the accuracy of the classification. The theoretical part also explains the influence of seasonality and weather on crop classification and analyzes the impact of soil properties on the spectral characteristics of crops. The practical part focuses on specific data, methods of processing, and the analysis of results, including the determination of crop growing seasons.