Classification of Sentinel-2 Images Utilizing Abundance Representation

Abstract

This paper deals with (both supervised and unsupervised) classification of multispectral Sentinel-2 images, utilizing the abundance representation of the pixels of interest. The latter pixel representation uncovers the hidden structured regions that are not often available in the reference maps. Additionally, it encourages class distinctions and bolsters accuracy. The adopted methodology, which has been successfully applied to hyperpsectral data, involves two main stages: (I) the determination of the pixel’s abundance representation; and (II) the employment of a classification algorithm applied to the abundance representations. More specifically, stage (I) incorporates two key processes, namely (a) endmember extraction, utilizing spectrally homogeneous regions of interest (ROIs); and (b) spectral unmixing, which hinges upon the endmember selection. The adopted spectral unmixing process assumes the linear mixing model (LMM), where each pixel is expressed as a linear combination of the endmembers. The pixel’s abundance vector is estimated via a variational Bayes algorithm that is based on a suitably defined hierarchical Bayesian model. The resulting abundance vectors are then fed to stage (II), where two off-the-shelf supervised classification approaches (namely nearest neighbor (NN) classification and support vector machines (SVM)), as well as an unsupervised classification process (namely the online adaptive possibilistic c-means (OAPCM) clustering algorithm), are adopted. Experiments are performed on a Sentinel-2 image acquired for a specific region of the Northern Pindos National Park in north-western Greece containing water, vegetation and bare soil areas. The experimental results demonstrate that the ad-hoc classification approaches utilizing abundance representations of the pixels outperform those utilizing the spectral signatures of the pixels in terms of accuracy.

Fluvial gravel bar mapping with spectral signal mixture analysis

“…For water, the reflectance values of the automatically selected endmembers are sometimes higher (as in the case of Landsat 8) and sometime lower (Sentinel-2 2015) than the values of the manually selected endmembers. Spectral signature shapes are similar to those reported in the literature both for Landsat (Afrasinei et al, 2018;Wu, 2004) and for Sentinel-2 (Mylona et al, 2018). The spectral signatures of gravel are very similar to those of built-up land from the literature, while forest (vegetation 2) is less bright than reported in similar studies (Xi et al, 2019), possibly due to terrain shadow.…”

“…Since then, it has been used for various purposes, including land cover mapping (Ling et al, 2016), determining land cover fractions in urban areas (Kärdi, 2007), soil degradation monitoring (Dubovyk et al, 2015), grassland monitoring (Shao et al, 2018), river bank mapping (Niroumand-Jadidi & Vitti, 2017), and coastline mapping (Foody et al, 2005;Muslim et al, 2007). The SSMA has been used to analyse hyperspectral (Keshava, 2003;Somers et al, 2011) and multispectral satellite imagery, including Landsat (Wu, 2004) and Sentinel-2 (Mylona et al, 2018).…”

“…The SSMA determines the proportion of the selected land cover classes present within each pixel based on the pixel's values in different wavelengths. Thus, a sub-pixel mapping accuracy can be achieved (e.g., Ling et al, 2016;Mylona et al, 2018). The key information for the SSMA is the spectral reflectance of pure pixels (endmembers), i.e.…”

Sentinel-2 imagery usage on environmental monitoring of land use and occupation in a microwatershed in Central Brazil

“…Moreover, land-use and occupation can assist the socioeconomic and environmental planning of a given region (Novais et al 2016). Remote Sensing (SR) technologies associated with Geographic Information Systems (GIS) have provided identification of potentially productive lands or those most susceptible to environmental degradation (Radoux et al 2016;Borràs et al 2017;Mylona et al 2018;Kobayashi et al 2019;Nguyen et al 2020;Leopoldo et al 2020).…”