ARTMO’s MLCA toolbox: Machine Learning Classification Algorithms

Author: Jochem Verrelst (Jochem.verrelst @ uv.es) Presented: October 6, 2022 ARTMO’s MLCA toolbox: Machine Learning Classification Algorithms for automated evaluation of classifiers and thematic mapping The focus of today’s tutorial will be on ARTMO’s (Automated Radiative Transfer Models Operator) Machine Learning Classification Algorithms (MLCA) toolbox and some application examples. Satellite images provide valuable geospatial data for monitoring land surface conditions and characterizing and mapping land use/land cover. In this context, accurate plant types (PTs) detection forms an important basis for sustainable land management maintaining biodiversity and ecosystem services. Most importantly, the selection of best-performing algorithms needs to be considered for obtaining as accurate as possible PTs classification. So far, no freely downloadable toolbox exists that brings the diversity of the latest supervised machine learning classification algorithms (MLCAs) together into a single intuitive user-friendly graphical user interface (GUI). To fill this gap and to facilitate and automate the usage of MLCAs, here we present a novel GUI software package that allows systematically training, validating, and applying pixel based MLCA models to remote sensing imagery. The so-called MLCA toolbox has been integrated within ARTMO's software framework developed in MATLAB, which includes most of the state-of-the-art methods of the machine learning community. Furthermore, together with the MLCA toolbox, also the so-called LabelMeClass tool has been released, which facilitates the collection of labeled data used as input into the MLCA toolbox. The presentation will be composed of an introduction (brief theoretical session) and a practical demonstration part with the LabelMeClass tool and MLCA ARTMO toolbox. The full software framework can be freely downloaded at https://artmotoolbox.com/.