Comparing forest restoration canopy cover measurements using RGB and multispectral sensors onboard drones

Remotely Piloted Aircrafts (RPA) coupled with Red-Green-Blue (RGB) sensors have a high potential to monitor Forest Restoration (FR), but multispectral sensors onboard RPA are more expensive and still demand more studies when applied to FR monitoring. This work aims to compare an RGB and a multispectral sensor capacity to measure the canopy cover of a FR project. Four canopy cover methods were evaluated using: the point cloud data generated by the RGB sensor; a vegetation index for RGB sensors; the Normalized Difference Vegetation Index (NDVI); and the Near Infra-Red band (Nir) only. The point cloud data method was the most accurate and the only one that presented all accuracies greater than 0.9. However, the multispectral sensor presented more potential for scientific research because it seems to be capable of detecting different photosynthetic activities on the trees and, consequently, different responses to FR treatments, which should be confirmed by future studies.