Applying Reject Region to Adaptive Feature Extraction for Hyperspectral Image Classification

Abstract

In this study, a novel classifier ensemble method named adaptive feature extraction (AdaFE) with reject region is proposed for hyperspectral image. This new concept is deduced from the concepts of reject region and feature extraction. The main idea is adaptive in the sense that subsequent feature spaces are tweaked in favor of those reject regions by Gaussian or knn classifiers in the previous feature space. All training samples are projected to these feature spaces to train various classifiers and then constitute a multiple classifier system. The experimental results based on two hyperspectral data sets show that the proposed algorithm can generate better classification results than only applying feature extraction.