Robert is a PhD candidate at the Laboratory of Geo-information Science and Remote sensing of Wageningen University, working on his PhD research topic focusing on assessing land use change on national and sub-national scales with remote sensing time series. The PhD is part of the CIFOR's global comparative study on REDD + (GCS).
He intends to exploit the advancement in earth observation and artificial intelligence to enable the detection of spatial patterns in environmental data based on satellite images. The detection of varying spatial patterns is critical for (1) land use characterization at small scale and larger scale, (2) time series land use change dynamics assessment, and (3) monitoring of proximal drivers of land use change associated with forest carbon emissions. The use of Deep learning is rather essential for the automated assessment of the interacting and complex land use types (natural forest vs plantation forest, Commercial agriculture vs Small scale agriculture Vs Pasture). Apart from spatial land use pattern characterization. He will also incorporate time series stack of Sentinel -1, Sentinel -2 and Landsat images into deep learning algorithms for time series analysis. The algorithms will be trained by the visually interpreted land use patches.
Robert holds a Master of Science degree in Geo-information Science and Earth Observation from the Faculty of Geo-information science and earth observation of the University of Twente with a research specialization in using remote sensing data for aboveground biomass and carbon stock assessment. He has +3 years of experience working in forest resource assessment and conservation for the Tanzanian Forest Service.
He enjoys carefully thinking about the highly mathematical parts of analyzing natural resources problems, particularly land use characterization and land use dynamics assessment, with the application of spatial statistical modeling, machine learning, and deep learning algorithms.