In Geographic Object-based Image Analysis (GEOBIA), remote sensing experts benefit from a large spectrum of characteristics to interpret images (spectral information, texture, geometry, spatial relations, etc). However, the quality of a classification is not always increased by considering a higher number of features. The experts are then used to define classification rules based on a laborious "trial-and-error" process. In this paper, we test a methodology to automatically determine an optimal subset of features for discriminating features. This method assumes that a reference land cover map (or at least training samples) is available. Two approaches were considered: a rule-based approach and a Support Vector Machine approach. For each approach, the method consists in ranking the features according to their potential for discriminating two classes. This task was performed thanks to the Jeffries-Matusita distance and Support Vector Machine-Ranking Feature Extraction (SVM-RFE) algorithm. Then, it consists in training and validating a classification algorithm (rule-based and SVM), with an increasing number of features: first only the best-ranked feature is included in the classifier, then the two best-ranked features, etc., until all the N features are included. The objective is to analyze how the quality of the classification evolves according to the numbers of features used. The optimal subset of features is finally determined through the analysis of the Akaike information criterion. The methodology was tested on two classes (urban an non urban areas) on a Spot5 image regarding a study area located in the La Réunion island.