Input Images for the article : "Self-supervised learning based clustering workflow for exploring seismological data from dense networks"

This repository contains a dataset of 1,966,581 images representing 30-second waveform segments and their corresponding spectrograms, derived from the continuous seismic record of the "Marie-sur-Tinée" landslide dataset. The "Marie-sur-Tinée" landslide dataset was recorded from April 18, 2021, to June 28, 2021, and includes data from 69 seismic stations. The 1.9 million images generated from the data of 13 stations were used to train a Self-Supervised Learning (SSL) model which was later used to encode the entire dataset. The clustering results based on the embedding returned by the model are presented in the article (submitted): Rimpot J., Hibert C., Malet J.-P., Forestier G., and Weber J., Self-supervised Learning Based Clustering Workflow for Exploring Seismological Data from Dense Networks