Charles Explorer logo
🇨🇿

Unmanned aircraft in nature conservation: an example from plant invasions

Publikace na Přírodovědecká fakulta |
2017

Tento text není v aktuálním jazyce dostupný. Zobrazuje se verze "en".Abstrakt

To successfully fight plant invasions, new methods enabling fast and efficient monitoring are needed, and remote sensing can make their management more efficient and less expensive. Optimal solution depends on the species characteristics, where the spectral and spatial resolution can compensate each other to some extent, and phenology plays an important role.

Available high spatial resolution satellite data are sufficient for recognition of species that are distinct and either large or form uniform patches at size comparable to the data pixel size. For other species, higher spatial resolution is needed, and unmanned aircraft (UAV) provide data of extremely high spatial resolution (cm) at low cost and high flexibility.

We assess its potential to map invasive black locust (BL, Robinia pseudoaccacia), testing imagery of different origin (satellite, UAV), spectral (multispectral, red, green, and blue (RGB) + near-infrared (NIR)) and spatial resolution, and various technical approaches to choose the best strategy for the species monitoring balancing between precision of detection and economic feasibility. Using purposely designed low-cost UAV with tailless fixed wing design for two consumer cameras (RGB and modified NIR) ensures robustness and repeatable field performance while maintaining high aerodynamic efficiency, with resulting mapping capacity over 10 km(2) per day.

For repeated measurements, it is extremely important to ensure spatial co-registration of pixels/objects from different phenological phases. Investment in GPS receiver in the UAV and GPS post-processing eliminated laborious collection of ground control points, while maintaining the co-registration of objects across multiple flights.

In our study we provide evidence of benefit of the low cost unmanned system for species monitoring with high classification accuracies of target species from UAV orthomosaic outcompeting WorldView-2 satellite data.