Original scientific paper
https://doi.org/10.5513/JCEA01/25.3.4278
Vegetation mapping based on visual data
Kristóf KOZMA-BOGNÁR
; Hungarian University of Agriculture and Life Sciences, Georgikon Campus, H-8360 Keszthely, Deák Ferenc str. 16., Hungary
*
József BERKE
; Dennis Gabor University, H-1119 Budapest, Fejér Lipót str. 70., Hungary
Angéla ANDA
; Hungarian University of Agriculture and Life Sciences, Georgikon Campus, H-8360 Keszthely, Deák Ferenc str. 16., Hungary
Veronika KOZMA-BOGNÁR
; Dennis Gabor University, H-1119 Budapest, Fejér Lipót str. 70., Hungary
* Corresponding author.
Abstract
This research aimed to present the database construction process using some plant species found in the sample area. In addition to aerial photography, the characteristic plant species were determined through fieldwork, and their coordinates were recorded with the iPoint software. In the course of the research, DJI Phantom 4 Pro and Mavic Mini drones were used to take aerial photographs of Zimányi-island in Kis-Balaton. Several photographs were taken of the individual plants for identification and assessment of association characteristics, thereby broadening the spectrum of processed data. The aerial photos obtained during the recordings were initially processed with Agisoft PhotoScan 1.4.3 (Metashape after version 1.5), then for comparison with Agisoft Metashape 2.0.3 and Agisoft Cloud software. By combining the aerial photos, a high-resolution TIFF orthophoto was obtained for the entire area. After that, the visual data sets required for the 3D model-based vegetation map were created, pre-processed and processed. The integration of visual and non-visual data sets were integrated into the GIS system. In the research area, non-image data were also collected, which were entered into Microsoft Office Excel tables. QGIS software was used to insert these tables into the database. The completed project shows the coordinates of the 27 plant species recorded in 2019 and the 28 plant species recorded in 2020. The process of querying the developed database was also presented. Finally, a comparative analysis of two different versions of the software was presented and used for matching in the past period. The method can be widely used in precision agriculture as well, such as weed detection, which is pivotal to increasing crop yields.
Keywords
GIS; remote sensing; drone; environmental protection
Hrčak ID:
320924
URI
Publication date:
23.9.2024.
Visits: 140 *