Country-wide high-resolution vegetation height mapping with Sentinel-2

Amsterdam

2019

1-15

Sonderdruck

204786

Sentinel-2 multi-spectral images collected over periods of several months were used to estimate vegetation
height for Gabon and Switzerland. A deep convolutional neural network (CNN) was trained to extract suitable
spectral and textural features from reflectance images and to regress per-pixel vegetation height. In Gabon,
reference heights for training and validation were derived from airborne LiDAR measurements. In Switzerland,
reference heights were taken from an existing canopy height model derived via photogrammetric surface reconstruction. The resulting maps have a mean absolute error (MAE) of 1.7m in Switzerland and 4.3m in Gabon
(a root mean square error (RMSE) of 3.4m and 5.6 m, respectively), and correctly estimate vegetation heights up
to>50 m. They also show good qualitative agreement with existing vegetation height maps. Our work demonstrates
that, given a moderate amount of reference data (i.e., 2000 km2 in Gabon and ≈5800 km2 in Switzerland), high-resolution vegetation height maps with 10m ground sampling distance (GSD) can be derived
at country scale from Sentinel-2 imagery.
Keywords: Vegetation height mapping, Convolutional neural network, Deep learning, Sentinel-2