Comparison of optical sensors discrimination ability using spectral libraries

Abstract

In remote sensing, the ability to discriminate different land covers or material types is directly linked with the spectral resolution and sampling provided by the optical sensor. Previous studies have shown that spectral resolution is a critical issue, especially in complex environments. In spite of the increasing availability of hyperspectral data, multispectral optical sensors onboard various satellites are acquiring every day a massive amount of data with a relatively poor spectral resolution (i.e. usually about four to seven spectral bands). These remotely sensed data are intensively used for Earth observation regardless of their limited spectral resolution. In this article, we studied seven of these optical sensors: Pleiades, QuickBird, Système Pour l'Observation de la Terre 5 (SPOT5), IKONOS, Landsat Thematic Mapper (TM), FORMOSAT, and Medium Resolution Imaging Spectrometer (MERIS). This study focuses on the ability of each sensor to discriminate different materials according to its spectral resolution. We used four different spectral libraries which contain around 2500 spectra of materials and land covers with a fine spectral resolution. These spectra were convolved with the relative spectral responses (RSRs) of each sensor to create spectra at the sensors' resolutions. Then, these reduced spectra were compared using separability indices (divergence, transformed divergence (TD), Bhattacharyya, and Jeffreys-Matusita) and machine learning tools. In the experiments, we highlighted that the configuration of spectral bands could lead to important differences in classification accuracy according to the context of application (e.g. urban area).