A new hierarchical classifier for hyperspctral data with similar spectrum

To improve the classification accuracy of image in which many classes have the similar spectrum, this paper presents a new hierarchical classification scheme for hyperspectral images (HSI). The Spectral Angle Mapping (SAM) is firstly used to combine the similar classes into large classes. Next the hierarchical classifier classifies the image with large classes and then divides every large class into normal classes further. For every large class, the most suitable feature extraction method and classifier are chosen empirically. Meanwhile, a new band selection is proposed to help every large class find the bands which can better reflect the differences of classes according to the characteristic of spectrum. Experiments are conducted on a 103-band ROSIS image of University of Pavia. The experimental results show that the hierarchical classifier is better than the single classifier used only once. Especially when the spectra of the given classes are so similar that the traditional classifiers couldn't divide them thoroughly, the proposed classifier can make it. Moreover, the hierarchical classifier can do more efficiently because it excludes some redundant bands and concentrates on the bands with slight differences.

1. INTRODUCTION

Over the past few years, comprehensive systems of classification, such as decision tree and hierarchical classifier, have been widely applied in classification of remote sensing images [1]. For example, Monteiro and Murphy [2] took into account the improvement of each split of the tree ensemble and calculated a relative measure of the input information. Li et.al.[3] adopted pure pixel index (PPI) to extract endmembers as training samples and used C4.5 decision tree algorithm to classify. Bakos and Gamba [4] introduced a novel methodology to build a multistage hierarchical data processing approach that is able to combine the advantages of different processing chains, which may be best suited for special classes. Jensen et.al.[5] introduced a framework for interpreting a broad range of such regularization matrices in the linear and quadratic discriminant analysis (LDA and QDA, respectively) classifier settings. Suju et.al.[6] used the binary hierarchical classifier to propose a knowledge transfer framework that leverages the information extracted from the existing labeled data to classify spatially separate and multi-temporal test data. Bruzzone et.al.[7] researched the issue of bidirectional reflectance and pointed out the naive use of a classifier which is trained on the available ground-truth data from one region on data that are from spatially or temporally different areas without accounting for the variability of the class signatures, would result in poor classification accuracies.