Clustering Analysis for Cloud and Surface Type Classification
Collaborator: Paul Menzel (Senior Scientist, University of Wisconsin, Verner Suomi Distinguished Professor)
We are currently developing a library which implements a range of clustering algorithms, some classification algorithms and functions to create simulated data from probability densities. The documentation for this library is available here
We have applied clustering algorithms to multi-modal MODIS level 1B data to automatically separate surface features such as snow and cloud types. This is aimed at finding a minimal channel set for these tasks.
Poster: Clustering Poster