Faculty of Mathematics and Natural Sciences - Earth Observation Lab

Faculty of Mathematics and Natural Sciences | Geography Department | Earth Observation Lab | News | Archive | New paper published in Remote Sensing on "Estimating Fractional Shrub Cover Using Simulated EnMAP Data: a Comparison of Three Machine Learning Regression Techniques" by Marcel Schwieder et. al.

New paper published in Remote Sensing on "Estimating Fractional Shrub Cover Using Simulated EnMAP Data: a Comparison of Three Machine Learning Regression Techniques" by Marcel Schwieder et. al.

In this research we used simulated hyperspectral EnMAP data for sub-pixel mapping of shrub cover in southern Portugal. Therefore, we compared the prediction performance of Support Vector Regression, Random Forest Regression and Partial Least Squares Regression...

Estimating Fractional Shrub Cover Using Simulated EnMAP Data: a Comparison of Three Machine Learning Regression Techniques

In this research we used simulated hyperspectral EnMAP data for sub-pixel mapping of shrub cover in southern Portugal. Therefore, we compared the prediction performance of Support Vector Regression, Random Forest Regression and Partial Least Squares Regression. In our tests SVR showed the best results, followed by RF and PLSR. We used the SVR model to produce a fractional shrub cover map for our study area, which revealed comprehensible patterns of shrubs. These patterns were particularly pronounced between regions with special land management incentives and those regions that were mainly left unmanaged. We showed that EnMAP data are valuable for detailed mapping of gradients in natural to semi-natural ecosystems and that machine learning algorithms enable us to do that in an efficient way.