Band Priority Index: A Feature Selection Framework for Hyperspectral Imagery

Wenqiang Zhang,Liaoying Zhao,Xiaorun Li

doi:10.3390/rs10071095

Wenqiang Zhang, Liaoying Zhao + Show 1 more

Open Access

https://doi.org/10.3390/rs10071095

Copy DOI

Journal: Remote sensing	Publication Date: Jul 10, 2018
Citations: 14	License type: CC BY 4.0

Affiliation: Zhejiang University, Hangzhou Dianzi University

Abstract

Hyperspectral Band Selection (BS) aims to select a few informative and distinctive bands to represent the whole image cube. In this paper, an unsupervised BS framework named the band priority index (BPI) is proposed. The basic idea of BPI is to find the bands with large amounts of information and low correlation. Sequential forward search (SFS) is used to avoid an exhaustive search, and the objective function of BPI consist of two parts: the information metric and the correlation metric. We proposed a new band correlation metric, namely, the joint correlation coefficient (JCC), to estimate the joint correlation between a single band and multiple bands. JCC uses the angle between a band and the hyperplane determined by a band set to evaluate the correlation between them. To estimate the amount of information, the variance and entropy are used as the information metric for BPI, respectively. Since BPI is a framework for BS, other information metrics and different mathematic functions of the angle can also be used in the model, which means there are various implementations of BPI. The BPI-based methods have the advantages as follows: (1) The selected bands are informative and distinctive. (2) The BPI-based methods usually have good computational efficiencies. (3) These methods have the potential to determine the number of bands to be selected. The experimental results on different real hyperspectral datasets demonstrate that the BPI-based methods are highly efficient and accurate BS methods.

Highlights

Hyperspectral images contain hundreds of bands with a fine resolution, e.g., 0.01 μm, which makes it possible to reduce overlap between classes, and, enhances the potential to discriminate subtle spectral difference [1,2]
In this paper, we propose a new Band Selection (BS) approach, named the Band Priority Index (BPI), which is a model or framework for BS and different metrics can be used in the model
A new objective function is designed for band priority index (BPI), and it consists of two parts: the information metric and the correlation metric

Summary

Introduction

Hyperspectral images contain hundreds of bands with a fine resolution, e.g., 0.01 μm, which makes it possible to reduce overlap between classes, and, enhances the potential to discriminate subtle spectral difference [1,2]. The high resolution of the spectrum makes the bands highly correlated. To process data effectively, dimensionality reduction (DR) is important and necessary. Dimensionality reduction techniques can be broadly split into two categories: feature extraction and feature selection (i.e., band selection) [3,4]. Feature extraction reduces the data dimensionality by extracting a set of new features from the original ones through some function mapping. The feature selection reduces the feature space by selecting a subset of features from the original features.

Objectives

Methods

Conclusion