Graph-based micro-seismic signal classification with an optimised feature space

Research output: Contribution to conferencePaper

10 Downloads (Pure)

Abstract

Classification of seismic events detected from seismic recordings has been gaining popularity in interpretation of subsurface processes, e.g., volcanic systems, earthquake activity, induced seismicity and slope stability, in particular landslides. However, due to the variability of signal representation for different classes in the temporal and spectral space, a large feature space to characterise the uniqueness of a particular type of event is used for classifying seismic signals. The consequence is additional complexity on the classifier and overfitting. So far, there has been little attempt to address dimensionality reduction via feature selection. In this paper, we propose an iterative, alternating graph feature and classifier learning method for micro-seismic signals via graph Laplacian regularization and normalized graph Laplacian regularization. Using recorded micro-seismic events from an active landslide, we demonstrate improved classification accuracy with a relatively small feature space compared to state of the art.
Original languageEnglish
Number of pages4
Publication statusPublished - 24 Jul 2020
Event2020 IEEE International Geoscience and Remote Sensing Symposium - Waikoloa, United States
Duration: 19 Jul 202024 Jul 2020
https://igarss2020.org/default.asp

Conference

Conference2020 IEEE International Geoscience and Remote Sensing Symposium
Abbreviated titleIEEE IGARSS 2020
CountryUnited States
CityWaikoloa
Period19/07/2024/07/20
Internet address

Keywords

  • micro-seismic
  • graph feature learning
  • GLR
  • norm-GLR

Fingerprint Dive into the research topics of 'Graph-based micro-seismic signal classification with an optimised feature space'. Together they form a unique fingerprint.

  • Projects

    Cite this

    Li, J., Yang, C., Stankovic, V., Stankovic, L., & Pytharouli, S. (2020). Graph-based micro-seismic signal classification with an optimised feature space. Paper presented at 2020 IEEE International Geoscience and Remote Sensing Symposium, Waikoloa, United States.