Ensemble classifiers in remote sensing: A review

An ensemble consists of a set of individually trained classifiers (e.g., as neural networks or decision trees etc.) whose predictions are combined in some manner (e.g., averaging or voting etc.) to form the final prediction. In literature many previous study has shown that an ensemble is often more accurate than any of the single classifiers. Ensemble learning is primarily used to improve the (classification, prediction, function approximation, etc.) performance of a model, or in other words it reduces the likelihood of the selection of a poor classifier or model. This paper focus on the fundamental concept of ensemble classifiers reviews the different ensemble techniques (Boosting, Bagging, and Random Forest) especially useful for remote sensing. Random Forest is one of the most popular ensemble classifier. In this paper, more emphasis is given to RF classifier and various application domain of this ensemble classifier for satellite image processing. This article serves as a foundation as well as provides a guideline for the new researcher in order to explore the applicability of RF classifier the area of remote sensing.