Source Separation for Simultaneous Seismic Data Acquisition

Warning message

You must login to view this form.
Submission Dates:
07/03/2017 to 09/01/2017
Citation Author(s):
IEEE MLSP Technical Committee
Submitted by:
Yuejie Chi
Last updated:
Tue, 05/01/2018 - 15:07
DOI:
10.21227/H2TP46
Data Format:
License:
Creative Commons Attribution

Abstract 

This is the data competion hosted by the IEEE Machine Learning for Signal Processing (MLSP) Technical Committee as part of the 27th IEEE International Workshop on Machine Learning for Signal Processing (MLSP 2017), Tokyo, Japan. This year the competion is based on a dataset kindly provided Petroleum Geo-Systems (PGS), on source separation for seismic data acquistion. 

The best three contenders are invited to present their methods and results at the IEEE Workshop on Machine Learning for Signal Processing that will be held in Tokyo this year from the 25th to the 28th September. A prize will also be awarded at this event. As part of the incentive, the organizing committee waives the conference registration fees for the top three contenders.

Instructions: 

Participants are asked develop deblending or source separation techniques to remove the effect of interfering sources from the data matrix. A sample dataset of a marine survey, PGS Triple Source data, has been provided by Petroleum Geo-Systems (PGS) for the purposes of this challenge. Please refer to the documentation for the detailed description of the data competition. 

Please contact Karim Seghouane <abd-krim.seghouane@unimelb.edu.au> and Yuejie Chi <chi.97@osu.edu> if you have questions regarding this competition.

Competition Dataset Files

AttachmentSize
File Time delay matrix1.54 KB
File blended_data.mat752.15 MB
    ACCESS ON AWS

    Documentation

    AttachmentSize
    File 2017_SeismicDeblendingSummary_Final.pdf876.33 KB