Autoencoder-based feature extraction for the automatic detection of snow avalanches in seismic data

Avalanche seismic detection systems are key for forecasting, but distinguishing avalanches from other seismic sources remains challenging. We propose novel autoencoder models to automatically extract features and compare them with a standard seismic feature extractor. These features are then used to classify avalanches and noise events. The autoencoder feature classifiers have the highest sensitivity to detect avalanches, while the standard seismic classifier performs better overall.