The large volume of RS data, the complexity of the landscape in a study area, as well as limited and usually imbalanced training data, make the classification a challenging task. Efficiency and computational cost of RS image classification is also influenced by different factors, such as classification algorithms, sensor types, training samples, input features, pre- and post-processing techniques, ancillary data, target classes, and the accuracy of the final product. Accordingly, these factors should be considered with caution for improving the accuracy of the final classification map. Carrying a simple accuracy assessment, through the Overall Accuracy (OA) and Kappa coefficient of agreement (K), by the inclusion of ground truth data might be the most common and reliable approach for reporting the accuracy of thematic maps. These accuracy measures make the classification algorithms comparable when independent training and validation data are incorporated into the classification scheme. Given the development and employment of new classification algorithms, several review articles have been published. To date, most of these reviews on remote sensing classification algorithms have provided useful guidelines on the general characteristics of a large group of techniques and methodologies.