We perform experiments with six popular ConvNets using three remote sensing datasets.
The primary emphasis of the program will remain on using remote sensing datasets for LCLUC research.
Acquired global remote sensing datasets require ground reference data in order to calibrate them and assess their quality.
In order to establish such a long-term continuous monitoring of diverse regions around the world, time series analyses of remote sensing datasets time series approaches are required.
The three remote sensing datasets show generally similar but distinctive performances in characterizing the two key variables over various land cover types.
Recently, increasing availability of remote sensing datasets with improved spatial accuracy, advances in computational power and geographical information system (GIS), enable evaluation of morphometric parameters with ease and better accuracy (Grohmann 2004).
This development has been catalyzed by the broad availability of improved remote sensing datasets and the evolution of cutting-edge data mining techniques for remote sensing data analysis [7, 14].
This is particularly timely considering several upcoming active remote sensing datasets that will be used to map forest biomass at a global scale (e.g. NASA's GEDI, NISAR, ESA's BIOMASS).
