The neural network using an unsupervised generalized Hebbian algorithm (GHA) is adopted to find the principal eigenvectors of a covariatice matrix in different kinds of seismograms. Prineipal components analysis (PCA) using (he neural net of GHA allows one to extract information of seismic reflection layers and uniform neighboring traces, for PCA. a theorem is proposed that adding extra poinl along the direction of the existing eigenvector can enhance that eigenvector. This theorem is applied to the Fault anal) sis and a reaJ seismogram al Mississippi Canyon. The results of PCA using neural network can improve the seismic interpretation.
|State||Published - Oct 2002|
|Event||2002 Society of Exploration Geophysicists Annual Meeting, SEG 2002 - Salt Lake City, United States|
Duration: 6 Oct 2002 → 11 Oct 2002
|Conference||2002 Society of Exploration Geophysicists Annual Meeting, SEG 2002|
|City||Salt Lake City|
|Period||6/10/02 → 11/10/02|