A Multiscale Random Forest Kernel for Land Cover Classification

Abstract

Random forest (RF) is a popular ensemble learning method that is widely used for the analysis of remote sensing images. RF also has connections with the kernel-based method. Its tree-based structure can generate an RF kernel (RFK) that provides an alternative to common kernels such as radial basis function (RBF) in kernel-based methods such as support vector machine (SVM). Using an RFK in an SVM has been shown to outperform both RF and SVM-RBF (i.e., using an RBF kernel in an SVM) in classification tasks with a high number of features. Here, we explore new designs of RFKs for remote sensing image classification. Different RF structural parameters and characteristics are used to generate various RFKs. In particular, we explore the use of RFs depth, the number of branches between terminal nodes of trees, and the predicted class probabilities for designing and evaluating new RFK. Two depth-based kernel are proposed: an RFK at the optimal depth, and a multiscale one created by combining RFKs at multiple depths. We evaluate the proposed kernels within an SVM by classifying a time series of Worldview-2 images, and by designing experiments having a various number of features. Benchmarking the new RFKs against the RBF shows that the newly proposed kernels outperform the RBF kernel for the experiments with a higher number of features. For the experiments with a lower number of features, RFKs and RBF kernel perform at about the same level. Benchmarking against the standard RF also shows the general outperformance of the proposed RFKs in an SVM. In all experiments, the best results are obtained with a depth-optimized RFK.