Class-Separation-Based Rotation Forest for Hyperspectral Image Classification

In this letter, we propose a new version of the rotation forest (RoF) method for the pixelwise classification of hyperspectral images. RoF, which is an ensemble of decision tree classifiers, uses random feature selection and data transformation techniques (i.e., principal component analysis) to improve both the accuracy of base classifiers and the diversity within the ensemble. Traditional RoF performs data transformation on the training samples of each subset. In order to further improve the performance of RoF, the data transformation is separately performed on each class, extracting sets of transformation matrices that are strictly dependent on the training samples of each single class. The approach, namely, class-separation-based RoF (RoF_CS), is experimentally investigated on a hyperspectral image collected by Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) sensor. Experimental results demonstrate that the proposed methodology achieves excellent performances, in comparison with random forest and RoF classifiers.

I. Introduction

In THE last few decades, hyperspectral image classification has been an incredibly active research topic with widespread applications [1]. However, classification of hyperspectral data is a challenge due to issues such as the high ratio of feature (spectral bands) to instance (training samples) and the redundant information in the feature set [2], [3]. In the past two decades, researchers have investigated a variety of approaches to alleviate such issues [4], [5].