Classification of Sentinel-2 Images Utilizing Abundances Representation

This paper deals with (both supervised and unsupervised) classification in Sentinel-2 images, utilizing the abundances representation of the pixels of interest. The latter pixel representation uncovers the hidden structured regions which are not available in the reference maps. Additionally, it reduces the dimensionality of the original spectral band space, it encourages data distinctions and bolsters accuracy. The proposed methodology involves two main stages: (I) the determination of the pixels abundance representation and (II) the employment of a classification algorithm applied on 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 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 the northwestern Greece containing water, vegetation and soil areas. The experimental results demonstrate that the ad-hoc classification approaches utilizing the abundance representations of the pixels outperform the ones utilizing the spectral signatures of the pixels, in terms of accuracy.