Historically in change detection, statistically based methods have been used. However, as the spatial resolutionof spectral images improves, the data no longer maintain a Gaussian distribution, and some assumptions aboutthe data - and subsequently all algorithms based upon those statistical assumptions - fail. Here we present theSimplex Volume Estimation algorithm (SVE), which avoids these potential hindrances by taking a geometricalapproach. In particular, we employ the linear mixture model to approximate the convex hull enclosing the datathrough identification of the simplex vertices (known as endmembers). SVE begins by processing an image andtiling it into squares. Next, SVE iterates through the tiles and for each set of pixels it identifies the numberof corners (as vectors) that define the simplex of that set of data. For each tile, it then iterates through theincreasing dimensionality, or number of endmembers, while every time calculating the volume of the simplexthat is defined by that number of endmembers. When the volume is calculated in a dimension that is higherthan that of the inherent dimensionality of the data, the volume will theoretically drop to zero. This value isindicative of the inherent dimensionality of the data as represented by the convex hull. Further, the volume ofthe simplex will fluctuate when a new material is introduced to the dataset, indicating a change in the image.The algorithm then analyzes the volume function associated with each tile and assigns the tile a metric valuebased on that function. The values of these metrics will be compared by using hyperspectral imagery collectedfrom different platforms over experimental setups with known changes between flights. Results from these testswill be presented along with a path forward for future research.
展开▼
