Biodiversity assessment using hierarchical clustering over hyperspectral images

Hyperspectral images represent an important source of information to assess ecosystem biodiversity. In particular, plant species richness is a primary indicator of biodiversity. This paper aims to use spectral variance to predict vegetation richness, known as Spectral Variation Hypothesis. A hierarchical clustering method based on minimum spanning tree computations retrieve clusters whose Shannon entropy reflects the species richness on a given zone. These entropies correlate well with the ones calculated directly from field data.