An Improved Endmember Selection Method Based on Vector Length for MODIS Reflectance Channels

Yuanliu Xu,Jinyang Du,Jiancheng Shi

doi:10.3390/rs70506280

Abstract

Endmember selection is the basis for sub-pixel land cover classifications using multiple endmember spectral mixture analysis (MESMA) that adopts variant endmember matrices for each pixel to mitigate errors caused by endmember variability in SMA. A spectral library covering a large number of endmembers can account for endmember variability, but it also lowers the computational efficiency. Therefore, an efficient endmember selection scheme to optimize the library is crucial to implement MESMA. In this study, we present an endmember selection method based on vector length. The spectra of a land cover class were divided into subsets using vector length intervals of the spectra, and the representative endmembers were derived from these subsets. Compared with the available endmember average RMSE (EAR) method, our approach improved the computational efficiency in endmember selection. The method accuracy was further evaluated using spectral libraries derived from the ground reference polygon and Moderate Resolution Imaging Spectroradiometer (MODIS) imagery respectively. Results using the different spectral libraries indicated that MESMA combined with the new approach performed slightly better than EAR method, with Kappa coefficient improved from 0.75 to 0.78. A MODIS image was used to test the mapping fraction, and the representative spectra based on vector length successfully modeled more than 90% spectra of the MODIS pixels by 2-endmember models.

Highlights

The accuracy of sub-pixel classification of land cover types using spectral mixture analysis (SMA)or multiple endmember spectral mixture analysis (MESMA) is strongly affected by the selection of pure spectra, or endmembers [1,2,3]
Endmember selection optimizing representative spectra from the endmember library is a crucial component for MESMA, as it balances the accuracy of modeled fractions and the computational efficiency of model fits [3,5]
Spectra derived from reference polygon were used to compare the new method to endmember average root mean squared error (RMSE) (EAR)/CoB method, and a Moderate Resolution Imaging Spectroradiometer (MODIS) image was used to test the performance of MESMA for mapping

Summary

Introduction

The accuracy of sub-pixel classification of land cover types using spectral mixture analysis (SMA)or multiple endmember spectral mixture analysis (MESMA) is strongly affected by the selection of pure spectra, or endmembers [1,2,3]. A count-based (CoB) method focused on the number of successful model fit within a library, and representative endmembers for each land cover class were chose with the spectra that successfully modeled the greatest number starting from all spectra to the spectra of a class remained unmodeled [2,6]. Another method, the minimum endmember average RMSE (EAR), based on the average error of a spectrum modeling all spectra of a class which was determined by MESMA, and the representative endmember is the minimum EAR spectrum within a class [3]. A combined method, hybrid IES-CoB/EAR selection, was developed to select endmembers for mapping plant, which synthesized the two previous approaches and successfully modeled plant species [5]

Objectives

Methods

Results

Discussion

Conclusion