The sustainability and protection of the olive trees is very important,as one of the oldest plant species in the Mediterranean. In recent decades, the development of precision agriculture methodology and geospatial technologies has made it possible to ensure sustainability in olive growing. Geospatial technologies enable the mapping and analysis of olive canopy in order to prevent various harmful pests and prolong lifetime. The research area of this graduation thesis is located in the northeastern part of the island of Pag, more precisely within the gardens of Lun olive groves, not far from the eponymous settlement Lun. The objectives of the research were to determine the positional accuracy of the UAV georeferencing process, to develop a multispectral model of very high resolution, to test the accuracy of classification algorithms, to analyze vegetation indices and to compare the approaches used. As a last goal, the variability of vegetation indices within olive canopies was determined. For the purpose of aerial photogrammetric recording of the investigated area, the Trinity F90+ UAV was used, and the recording was performed in two missions. The first mission involved the use of a very highresolution MicaSense RedEdge-MX Dual multispectral sensor, while the second mission involved the Sony RX1R II being attached to the drone to produce a very high-resolution DOP. Using multispectral models, the steps of the GEOBIA approach were performed, during which all user-defined parameters were compared and the most accurate classification algorithms were selected. The approach based on vegetation indices enabled the classification of canopies mainly due to the input data that include channels in the marginal red and infrared part of the electromagnetic spectrum. Olive canopies were selected using the VITO tool and then the accuracy of selected vegetation indices was examined using assessment measures. A comparison of the two approaches used was performed using five measures to assess accuracy. The last phase of the research was to determine the variability of vegetation indices within the canopy of individual olive trees. The positional accuracy of the obtained multispectral image is <3.5 cm. The Random Trees algorithm proved to be the best classification algorithm in the GEOBIA approach. After successful optimization of vegetation indices, the NDRE index was selected as the most accurate and representative example. Comparing the two approaches, the vegetation index-based approach showed better results in two methods and all five measures of accuracy, and was selected as the best in olive canopy mapping. Finally, the observed variability of vegetation indices within the canopy indicate the possibility of applying this methodology and represent the basis for future research in olive growing.