Cargando…
Applications of Unmanned Aerial Vehicle Based Imagery in Turfgrass Field Trials
Recent advances in remote sensing technology, especially in the area of Unmanned Aerial Vehicles (UAV) and Unmanned Aerial Systems (UASs) provide opportunities for turfgrass breeders to collect more comprehensive data during early stages of selection as well as in advanced trials. The goal of this s...
Autores principales: | , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Frontiers Media S.A.
2019
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6430071/ https://www.ncbi.nlm.nih.gov/pubmed/30930917 http://dx.doi.org/10.3389/fpls.2019.00279 |
Sumario: | Recent advances in remote sensing technology, especially in the area of Unmanned Aerial Vehicles (UAV) and Unmanned Aerial Systems (UASs) provide opportunities for turfgrass breeders to collect more comprehensive data during early stages of selection as well as in advanced trials. The goal of this study was to assess the use of UAV-based aerial imagery on replicated turfgrass field trials. Both visual (RGB) images and multispectral images were acquired with a small UAV platform on field trials of bermudagrass (Cynodon spp.) and zoysiagrass (Zoysia spp.) with plot sizes of 1.8 by 1.8 m and 0.9 by 0.9 m, respectively. Color indices and vegetation indices were calculated from the data extracted from UAV-based RGB images and multispectral images, respectively. Ground truth measurements including visual turfgrass quality, percent green cover, and normalized difference vegetation index (NDVI) were taken immediately following each UAV flight. Results from the study showed that ground-based NDVI can be predicted using UAV-based NDVI (R(2) = 0.90, RMSE = 0.03). Ground percent green cover can be predicted using both UAV-based NDVI (R(2) = 0.86, RMSE = 8.29) and visible atmospherically resistant index (VARI, R(2) = 0.87, RMSE = 7.77), warranting the use of the more affordable RGB camera to estimate ground percent green cover. Out of the top ten entries identified using ground measurements, 92% (12 out of 13 in bermudagrass) and 80% (9 out of 11 in zoysiagrass) overlapped with those using UAV-based imagery. These results suggest that UAV-based high-resolution imagery is a reliable and powerful tool for assessing turfgrass performance during variety trials. |
---|