Unsupervised Seismic Facies Classification Using Matching Pursuit and Self Organizing Maps

1 C016 Unsupervised Seismic Facies Classification Using Matching Pursuit and Self Organizing Maps MARCÍLIO C. MATOS 1 3 PAULO L.M. OSÓRIO 1 AND PAULO R.S. JOHANN 2 Abstract A new alternative to build seismic facies maps is presented. We propose to use matching pursuit with timefrequency dictionaries in each geological oriented segment of the temporal seismic trace associated with the Self Organizing Maps (SOM) as a clustering tool. Jointly they could be used as a seismic facies estimator. The technique was applied to a real data from a deep-water field in the Campos Basin Brazil. Introduction 1 Pontifícia Universidade Católica