HMM-based classification of seismic events recorded at Stromboli and Etna Volcanoes

摘要

This paper shows a complete seismic event classification system based on the state of the art in Hidden Markov Modeling (HMM) which has been successfully used for discriminating between different seismic events produced at active volcanoes. A database consisting of a representative set of different seismic events including explosions and tremor bursts recorded during active periods at Stromboli and Etna volcanoes in 1997 and 1999, respectively, was collected for training and testing. The method analyzes the seismograms comparing the characteristics of the data to a number of event classes defined beforehand. If a signal is present, the method detects its occurrence and produces a classification. The recognition and classification system based on HMM is a powerful, effective, and successful tool. From the application performed over our data set, we have demonstrated that in order to have a reliable result, a careful and adequate segmentation process is crucial.