RT info:eu-repo/semantics/article
T1 Plant survival monitoring with UAVs and multispectral data in difficult access afforested areas
A1 Gil-Docampo, M. L.
A1 Ortiz-Sanz, J.
A1 Martínez-Rodríguez, S.
A1 Marcos-Robles, J. L.
A1 Arza-García, M.
A1 Sánchez-Sastre, L. F.
K1 UAV
K1 plant control
K1 degraded region
K1 gap detection
K1 Waterboxx
AB Water supply devices enable afforestation in dry climates and on poor lands with generally high success rates. Previous survival analyses have been based on the direct observation of each individual plant in the field, which entails considerable effort and costs. This study provides a low-cost method to discriminate between live and dead plants in afforestation that can efficiently replace traditional field inspections through the use of unmanned aerial vehicles (UAVs) equipped with RGB and NIR sensors. The method combines the use of a conventional camera with an identical camera modified to record the NIR channel. Survival analysis was performed with digital image processing techniques based on calculated indices associated with plant vigour and PCA-based decorrelation. The method yielded results with high global accuracy rates (∼96.2%) with a minimum percentage of doubtful plants, even in young plantations (seedlings <30 cm tall). The procedure could be particularly useful in hazardous areas.
PB Taylor&Francis
SN 1010-6049
YR 2020
FD 2020
LK https://uvadoc.uva.es/handle/10324/81092
UL https://uvadoc.uva.es/handle/10324/81092
LA spa
NO Geocarto International, 2020, 35(2), 128–140.
NO Producción Científica
DS UVaDOC
RD 05-mar-2026