Spectral-Spatial Classification of Hyperspectral Image Based on Kernel Extreme Learning Machine

Chen Chen,Kui Liu,Hongjun Su,Wei Li

doi:10.3390/rs6065795

Abstract

Extreme learning machine (ELM) is a single-layer feedforward neural network based classifier that has attracted significant attention in computer vision and pattern recognition due to its fast learning speed and strong generalization. In this paper, we propose to integrate spectral-spatial information for hyperspectral image classification and exploit the benefits of using spatial features for the kernel based ELM (KELM) classifier. Specifically, Gabor filtering and multihypothesis (MH) prediction preprocessing are two approaches employed for spatial feature extraction. Gabor features have currently been successfully applied for hyperspectral image analysis due to the ability to represent useful spatial information. MH prediction preprocessing makes use of the spatial piecewise-continuous nature of hyperspectral imagery to integrate spectral and spatial information. The proposed Gabor-filtering-based KELM classifier and MH-prediction-based KELM classifier have been validated on two real hyperspectral datasets. Classification results demonstrate that the proposed methods outperform the conventional pixel-wise classifiers as well as Gabor-filtering-based support vector machine (SVM) and MH-prediction-based SVM in challenging small training sample size conditions.

Highlights

Hyperspectral imagery (HSI) captures reflectance values over a wide range of electromagnetic spectra for each pixel in the image
support vector machine (SVM)-based approaches have been extensively used for hyperspectral image classification since SVMs have often been found to outperform traditional statistical and neural methods, such as the maximum likelihood and the multilayer perceptron neural network classifiers [5]
The performance of the proposed spectral-spatial-based kernel based ELM (KELM) methods is shown in Tables 4 and 5 for two experimental data

Summary

Introduction

Hyperspectral imagery (HSI) captures reflectance values over a wide range of electromagnetic spectra for each pixel in the image. This rich spectral information allows for distinguishing or classifying materials with subtle differences in their reflectance signatures. Over the last two decades, many machine learning techniques including artificial neural networks (ANNs) and support vector machines (SVMs) have been successfully applied to hyperspectral image classification (e.g., [2,3,4,5]). SVMs have demonstrated excellent performance for classifying hyperspectral data when a relative low number of labeled training samples are available [4,5,7]. The SVM parameters (i.e., regularization and kernel parameters) have to be tuned for optimal classification performance

Methods

Results

Conclusion