search

stanford-futuredata / FAST

End-to-end earthquake detection pipeline via efficient time series similarity search
Apache License 2.0
145 stars 55 forks source link
earthquakes minhash-lsh-algorithm similarity-search time-series

readme

FAST

fast_index

FAST is an end-to-end and unsupervised earthquake detection pipeline. It is a useful tool for seismologists to extract more small earthquakes from continuous seismic data. FAST is able to run on different machines by using Google Colab, Linux, or Docker.

Check out the user guide to learn more about FAST and how to use it on your own dataset.

FAST User Guide Contents

  1. FAST Overview
    Click here for a summary of the FAST algorithm and why you might want to use it on your seismic data.

  2. Install
    Go here to learn how to install and run the FAST software on your computer.

    1. Google Colab

    2. Linux

    3. Docker

  3. Tutorial
    Learn how FAST detects earthquakes on the Hector Mine data set.

  4. How to Set Parameters
    Click here to learn how to test FAST on your own data sets.

    1. FAST Checklist

    2. Getting Seismic Data

    3. Input and Preprocessing

    4. Fingerprint

    5. Similarity Search

    6. Network Detection

    7. FAST Output

    8. Phase Picking

    9. Earthquake Location

    10. Example Parameters
      Click here to see data sets FAST has been used on to detect earthquakes.

  5. References
    Read publications about FAST here.

References

You can find more details about the pipeline and guidelines for setting parameters in our extended user guide. You may also check out the following papers: