Impact of feature selection on the accuracy and spatial uncertainty of per-field crop classification using Support Vector Machines

Crop mapping is one major component of agricultural resource monitoring using remote sensing. Yield orwater demand modeling requires that both, the total surface that is cultivated and the accurate distributionof crops, respectively is known. Map quality is crucial and influences the model outputs. Althoughthe use of multi-spectral time series data in crop mapping has been acknowledged, the potentially highdimensionality of the input data remains an issue. In this study Support Vector Machines (SVM) are usedfor crop classification in irrigated landscapes at the object-level. Input to the classifications is 71 multiseasonalspectral and geostatistical features computed from RapidEye time series. The random forest (RF)feature importance score was used to select a subset of features that achieved optimal accuracies. Therelationship between the hard result accuracy and the soft output from the SVM is investigated byemploying two measures of uncertainty, the maximum a posteriori probability and the alpha quadraticentropy. Specifically the effect of feature selection on map uncertainty is investigated by looking at thesoft outputs of the SVM, in addition to classical accuracy metrics. Overall the SVMs applied to the reduced feature subspaces that were composed of the most informative multi-seasonal features led to a clearincrease in classification accuracy up to 4.3%, and to a significant decline in thematic uncertainty. SVMwas shown to be affected by feature space size and could benefit from RF-based feature selection. Uncertaintymeasures from SVM are an informative source of information on the spatial distribution of error inthe crop maps