Skip to Main content Skip to Navigation
New interface

3D Seismic Attributes Enhancement and Detection by Advanced Technology of Image Analysis

Abstract : Moments have been extensively used in pattern recognition and image processing. In this thesis, we focus our attention on the study of 3D orthogonal Gaussian-Hermite moments, 2D and 3D Gaussian-Hermite moment invariants, fast algorithm of coherency attribute, and applications of seismic interpretation using moments methodology.We conduct seismic horizon auto-tracking methods from Gaussian-Hermite moments and moment invariants. We introduce multi-scale moment invariants approach. The experimental results show that method of 3D Gaussian-Hermite moments performs better than the most popular methods.We also approach seismic facies analysis based on feature vectors from 3D Gaussian-Hermite moments, and Self-Organizing Maps method with data visualization techniques. The excellent result shows that the integrated environment gives the best performance in interpreting the correct cluster structure.Finally, we introduce the parallel processing and volume visualization. Taking advantage of new performances by multi-threading and multi-cores technologies into seismic interpretation, we efficiently compute the seismic attributes and track the horizon. We also discuss volume rendering algorithm based on Open-Scene-Graph engine which provides better insight into the structure of seismic data.
Document type :
Complete list of metadata
Contributor : ABES STAR :  Contact
Submitted on : Thursday, September 13, 2012 - 4:54:18 PM
Last modification on : Tuesday, July 9, 2019 - 10:13:35 AM
Long-term archiving on: : Friday, December 16, 2016 - 12:50:38 PM


Version validated by the jury (STAR)


  • HAL Id : tel-00731886, version 1


Gengxiang Li. 3D Seismic Attributes Enhancement and Detection by Advanced Technology of Image Analysis. Earth Sciences. Université Michel de Montaigne - Bordeaux III, 2012. English. ⟨NNT : 2012BOR30006⟩. ⟨tel-00731886⟩



Record views


Files downloads