An integrated graph cuts segmentation and piece-wise convex unmixing approach for hyperspectral imaging

Context-based unmixing has been studied by several researchers. Recent techniques, such as piece-wise convex unmixing using fuzzy and possibilistic clustering or Bayesian methods proposed in [11] attempt to form contexts via clustering. It is assumed that the linear mixing model applies to each cluster (context) and endmembers and abundances are found for each cluster. As the clusters are spatially coherent, hyperspectral image segmentation can significantly aid unmixing approaches that perform cluster specific estimation of endmembers. In this work, we integrate a graph-cuts segmentation algorithm with piece-wise convex unmixing. This is compared to fuzzy clustering (FCM) with results obtained on two datasets. The results demonstrate that the integrated approach achieves better segmentation and more precise end-member identification (in terms of comparisons with known ground truth).