Compression-based unsupervised clustering of spectral signatures

Daniele Cerra, Jakub Bieniarz, Janja Avbelj, Peter Reinartz, R. Mueller

2011 · 11 citations · 14 references

Concepts

Abstract

This paper proposes to use compression-based similarity measures to cluster spectral signatures on the basis of their similarities. Such universal distances estimate the shared information between two objects by comparing their compression factors, which can be obtained by any standard compressor. Experiments on rocks categorization show that these methods may outperform traditional choices for spectral distances based on vector processing.

References

14