Semi-automatic interpretation of 3D seimentological Structures on Geological Images: an object-based approach

in: Interpretation, 3:3 (SX63-SX74)

Abstract

The characterization of sedimentary structures is an important step to construct quantitative models of sedimentary deposits from digital images such as 3D seismic data, satellite images or digital outcrops. However, the interpretation of these structures generally consists in tedious line pickings followed by surface modeling to define geobodies. Automatic geobody extraction is an alternative but it is sensitive to image noise and does not account for prior sedimentary knowledge. We propose to combine minimal picking by an interpreter with object-guided image processing and optimization to achieve fast and semi-automatic geobody interpretation. Our approach uses a realistic volumetric geobody representation based on Non-Uniform Rational B-Splines (NURBS) which can easily be deformed by the interpreter and numerical optimization. A custom edge detection guided by some initial rough interpretations is performed to strengthen the most relevant edges in the picture. An automatic optimization is then computed to fit the initial geobody to these highlighted edges. This approach is applied on satellite pictures showing alluvial channels and some preliminary results on 3D seismic time slices are also presented. The interpreted channels are then used in a retro-deformation process to automatically reconstruct the point bars. This semi-automatic method opens new perspectives to help interpreters rapidly come up with 3D models of sedimentary structures from subsurface and analog surface data sets.

Download / Links

    BibTeX Reference

    @ARTICLE{,
        author = { Ruiu, Jeremy and Caumon, Guillaume and Viseur, Sophie },
         title = { Semi-automatic interpretation of 3D seimentological Structures on Geological Images: an object-based approach },
         month = { "aug" },
       journal = { Interpretation },
        volume = { 3 },
        number = { 3 },
          year = { 2015 },
         pages = { SX63-SX74 },
      abstract = { The characterization of sedimentary structures is an important step to construct
    quantitative models of sedimentary deposits from digital images such as 3D
    seismic data, satellite images or digital outcrops. However, the interpretation
    of these structures generally consists in tedious line pickings followed by
    surface modeling to define geobodies. Automatic geobody extraction is an
    alternative but it is sensitive to image noise and does not account for prior
    sedimentary knowledge. We propose to combine minimal picking by an interpreter
    with object-guided image processing and optimization to achieve fast and
    semi-automatic geobody interpretation.  Our approach uses a realistic volumetric
    geobody representation based on Non-Uniform Rational B-Splines (NURBS) which can
    easily be deformed by the interpreter and numerical optimization. A custom edge
    detection guided by some initial rough interpretations is performed to
    strengthen the most relevant edges in the picture. An automatic optimization is
    then computed to fit the initial geobody to these highlighted edges. This
    approach is applied on satellite pictures showing alluvial channels and some
    preliminary results on 3D seismic time slices are also presented. The
    interpreted channels are then used in a retro-deformation process to
    automatically reconstruct the point bars. This semi-automatic method opens new
    perspectives to help interpreters rapidly come up with 3D models of sedimentary 
    structures from subsurface and analog surface data sets. }
    }