importance of wetlands, and unlike other critical ecosystems (cf. forests,
mangroves and inland water bodies), the extent and dynamics of wetland ecosystems
remains ill defined, characterized and modeled.
In this project
the aim would be to bridge this data gap by applying cutting edge machine
learning models to develop more accurate, consistent and comprehensive inventories
of the spatial extent of wetlands;
should explore the use deep learning to automatically detect and map wetlands
using earth observation satellite imagery. A combination of convolutional
neural network (CNN) and recurrent neural network (RNN) networks will be
applied to exploit the spatio temporal characteristics of wetlands.
Lydholm Rasmussen, firstname.lastname@example.org.
Birte Kronbak Andersen is not the
project supervisor, she just put the project idea in the Project Bank for
DHI Gras. But a supervisor can most likely be found at DTU Space.
I samarbejde medDHI Gras