subtraction method attenuates the noise better than the filtering approach Guitton and Claerbout (2003). This is difficult to answer...</SMALL>
<BR><B>Guitton A. and Claerbout J.</B> <BR><SMALL>We process a bathymetry survey from the Sea of Galilee. This
 HREF="http://sepwww.stanford.edu/data/media/public/docs/sep114/guojian2/paper_html/node5.html#GEO36-03-04670481">Claerbout (1971)</A> produces the correct phase 
 HREF="http://sepwww.stanford.edu/data/media/public/docs/sep114/bill1/paper_html/node6.html#gee">Claerbout (1999)</A>; <A NAME="tex2html23"
 HREF="http://sepwww.stanford.edu/data/media/public/docs/sep115/alejandro2/paper_html/node12.html#Claerbout.blackwell.85">Claerbout (1985)</A>. This approximation ignores back-scattering in the
 HREF="http://sepwww.stanford.edu/data/media/public/docs/sep115/alejandro2/paper_html/node12.html#GEO36.03.04670481">Claerbout (1971)</A>.
<BR><B>Guitton A., Claerbout J., and Lomask J.</B> <BR><SMALL>A new method for estimating interval velocities without 
<BR><B>Wolf K., Rosales D., Guitton A., and Claerbout J.</B> <BR><SMALL>At every point in a CMP gather, a local estimate of RMS velocity is:
 HREF="http://sepwww.stanford.edu/data/media/public/docs/sep117/bob2/paper_html/node12.html#gee">Claerbout (1999)</A>
<BR><B>Lomask J., Guitton A., Fomel S., and Claerbout J.</B> <BR><SMALL>We present a method for efficiently flattening 3D seismic data volumes.  First local dips are calculated over the entire seismic volume.  The dips are then resolved into time shifts using a Gauss-Newton iterative approach that exploits the Fourier domain to maximize efficiency.  To handle faults (discontinuous reflections), we apply a weight inversion scheme.  This approach successfully flattens a synthetic faulted model, a field salt peircement dataset, a field dataset with an angular unconformity, and a faulted field dataset.
 HREF="http://sepwww.stanford.edu/data/media/public/docs/sep120/bill2/paper_html/node5.html#Claerbout.blackwell.92">Claerbout (1992</A>,<A NAME="tex2html22"
 HREF="http://sepwww.stanford.edu/data/media/public/docs/sep124//jesse3/paper_html/node7.html#Lomask.sep.112.jesse1">Lomask and Claerbout (2002)</A>; <A NAME="tex2html25"
 HREF="http://sepwww.stanford.edu/data/media/private/docs/sep129/ben1/paper_html/node8.html#Claerbout.sep.67.139">Claerbout (1990)</A>, SEP progressed to reproducible
<h3>An algorithm for interpolation using Ronen's pyramid</h3>[<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep134/jon1/paper.pdf">pdf 103k</A>] [<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep134/jon1/jon1.tar.gz">src 0.7Mb</A>]<B>Jon Claerbout and Antoine Guitton</B> <BR><SMALL>
<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep136/joncrosstalk/paper_html/index.html">Anti-crosstalk</A> [<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep136/joncrosstalk/paper.pdf">pdf 76K</A>][<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep136/joncrosstalk/joncrosstalk.tar.gz">source</A>] <BR><B>Jon Claerbout</B> <BR><SMALL>Inverse theory can never be wrong because it's just theory.
<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep136/jonhubbert/paper_html/index.html">Hubbert math</A> [<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep136/jonhubbert/paper.pdf">pdf 76K</A>][<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep136/jonhubbert/jonhubbert.tar.gz">source</A>] <BR><B>Jon Claerbout and Francis Muir</B> <BR><SMALL>Hubbert fits growth and decay of petroleum production
<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep136/jontarsand/paper_html/index.html">Tar sands: Reprieve or apocalypse?</A> [<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep136/jontarsand/paper.pdf">pdf 436K</A>][<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep136/jontarsand/jontarsand.tar.gz">source</A>] <BR><B>Jon Claerbout</B> <BR><SMALL>Based on a Hubbert-type analysis
<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep139/jon1/paper_html/index.html">Blocky models via the L1/L2 hybrid norm</A> [<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep139/jon1/paper.pdf">pdf 112K</A>][<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep139/jon1/jon1.tar.gz">source</A>] <BR><B>Jon Claerbout</B> <BR><SMALL>This paper seeks to define robust,
<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep139/antoine1/paper_html/index.html">Theory and practice of interpolation in the pyramid domain</A> [<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep139/antoine1/paper.pdf">pdf 25M</A>][<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep139/antoine1/antoine1.tar.gz">source</A>] <BR><B>Antoine Guitton and Jon Claerbout</B> <BR><SMALL>With the pyramid transform, 2-D dip spectra can be characterized by
<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep140/yang1/paper_html/index.html">Least-squares imaging and deconvolution using the  hb norm conjugate-direction solver</A> [<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep140/yang1/paper.pdf">pdf 1.8M</A>][<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep140/yang1/yang1.tar.gz">source</A>] <BR><B>Yang Zhang and Jon Claerbout</B> <BR><SMALL>To retrieve a sparse model, we applied the hybrid norm
Claerbout to two interesting geophysical problems: least-squares imaging and blind deconvolution. The results showed that
<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep140/elita1/paper_html/index.html">Geophysical applications of a novel and robust L1 solver</A> [<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep140/elita1/paper.pdf">pdf 412K</A>][<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep140/elita1/elita1.tar.gz">source</A>] <BR><B>Yunyue Li, Yang Zhang, and Jon Claerbout</B> <BR><SMALL>L1-norm is better than L2-norm at dealing with noisy data and yielding
 HREF="elita1/paper_html/node8.html#Jon09">Claerbout (2009)</A> to perform L1 regressions. The solver is tested
<BR><B>Yang Zhang and Jon Claerbout</B><BR>
<BR><B>Chris Leader, Jon Claerbout, and Antoine Guitton</B> <BR>
<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep143/yishen/paper_html/index.html">A new algorithm for bidirectional deconvolution</A>[<a href="http://sepwww.stanford.edu/data/media/private/docs/sep143/yishen/paper.pdf"><img height=15 src=icons/pdf.png>paper</a>][<a href="http://sepwww.stanford.edu/data/media/private/docs/sep143/yishen/yishen.tar.gz"><img height=15 src=icons/GZIcon.png>source</a>] <BR><B>Yi Shen, Qiang Fu, and Jon Claerbout</B> <BR><SMALL>We introduce a new algorithm for bidirectional
 HREF="../sep142/index.html#Yang2010">Zhang and Claerbout (2010)</A>. 
<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep143/qiang/paper_html/index.html">An approximation of the inverse Ricker wavelet as an initial guess for bidirectional deconvolution</A> [<a href="http://sepwww.stanford.edu/data/media/private/docs/sep143/qiang/paper.pdf"><img height=15 src=icons/pdf.png>paper</a>][<a href="http://sepwww.stanford.edu/data/media/private/docs/sep143/qiang/qiang.tar.gz"><img height=15 src=icons/GZIcon.png>source</a>]<BR><B>Qiang Fu, Yi Shen, and Jon Claerbout</B> <BR><SMALL>Bidirectional deconvolution is a powerful tool for performing blind deconvolution on a signal that contains a mixed-phase wavelet, such as seismic data. Previously, we used a prediction error filter (PEF) as the minimum-phase filter and an impulse function as the maximum-phase filter as initial guesses. This surprised us by apparent instability as the solution would jump from spiking the first pulse of a Ricker wavelet to get larger middle pulse. In this paper, we propose new initial guesses for the causal and anti-causal deconvolution filters that are more effective on data with a Ricker-like wavelet. We test these on both synthetic data and field data. The results demonstrate that the new starting filters do a better job than the previous initial guesses.
<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep143/jon/paper_html/index.html">A log spectral approach to bidirectional deconvolution</A> [<a href="http://sepwww.stanford.edu/data/media/private/docs/sep143/jon/paper.pdf"><img height=15 src=icons/pdf.png>paper</a>][<a href="http://sepwww.stanford.edu/data/media/private/docs/sep143/jon/jon.tar.gz"><img height=15 src=icons/GZIcon.png>source</a>] <BR><B>Jon Claerbout, Qiang Fu, and Yi Shen</B> <BR><SMALL>The blind-deconvolution problem for non-minimum-phase-source
<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep145/qiang/paper_html/index.html">Data examples of logarithm Fourier-domain bidirectional deconvolution</A> [<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep145/qiang/paper.pdf">pdf 2.7M</A>]  [<a href="http://sepwww.stanford.edu/data/media/private/docs/sep145/qiang/qiang.tar.gz"><img height=15 src=icons/GZIcon.png>source</a>] <BR><B>Qiang Fu, Yi Shen, and Jon Claerbout</B>  <BR><SMALL>Time-domain bidirectional deconvolution methods show great promise for overcoming the minimum-phase assumption in blind deconvolution of signals containing a mixed-phase wavelet, such as seismic data. However, time-domain bidirectional methods usually suffer from slow convergence (Slalom method) or the starting model (Symmetric method). Claerbout proposed a logarithm Fourier-domain method to perform bidirectional deconvolution. In this paper, we test the new logarithm Fourier-domain method on both synthetic data and field data. The results demonstrate that the new method is more stable than previous methods and that it produces better results.
<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep145/yishen/paper_html/index.html">Preconditioning a non-linear problem and its application to bidirectional deconvolution</A> [<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep145/yishen/paper.pdf">pdf 1.5M</A>]  [<a href="http://sepwww.stanford.edu/data/media/private/docs/sep145/yishen/yishen.tar.gz"><img height=15 src=icons/GZIcon.png>source</a>] <BR><B>Yi Shen, Qiang Fu, and Jon Claerbout</B> <BR><SMALL>Non-linear optimization problems suffer from local minima. When we use gradient-based iterative solvers on these problems, we often find the final solution to be
<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep147/jon1/paper_html/index.html">Polarity preserving decon in ``N log N'' time</A> <a href="http://sepwww.stanford.edu/data/media/private/docs/sep147/jon1/paper.pdf"><img height=15 src=icons/pdf.png>PDF</a> <a href="http://sepwww.stanford.edu/data/media/private/docs/sep147/jon1/jon1.tar.gz"><img height=15 src=icons/GZIcon.png>source</a> <BR><B>Jon Claerbout</B> <BR><SMALL>A slight modification to Fourier spectral factorization
<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep147/jon2/paper_html/index.html">Decon in the log domain with variable gain</A> <a href="http://sepwww.stanford.edu/data/media/private/docs/sep147/jon2/paper.pdf"><img height=15 src=icons/pdf.png>PDF</a> <a href="http://sepwww.stanford.edu/data/media/private/docs/sep147/jon2/jon2.tar.gz"><img height=15 src=icons/GZIcon.png>source</a> <BR><B>Jon Claerbout, Antoine Guitton, and Qiang Fu</B> <BR><SMALL>We base deconvolution on the concept of output model sparsity.
<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep148/jon/paper_html/index.html">Modeling data error during deconvolution</A> [<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep148/jon/paper.pdf"><img height=15 src=icons/pdf.png>PDF 72K</A> <A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep148/jon/jon.tar.gz"><img height=15 src=icons/GZIcon.png>SRC</A>] <BR><B>Jon Claerbout and Antoine Guitton</B> <BR><SMALL>Our current decons take the data sacrosanct and find the best noncausal wavelet to deconvolve it with.
<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep148/antoine2/paper_html/index.html">Decon comparisons between Burg and conjugate-gradient methods</A> [<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep148/antoine2/paper.pdf"><img height=15 src=icons/pdf.png>PDF 408K</A> <A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep148/antoine2/antoine2.tar.gz"><img height=15 src=icons/GZIcon.png>SRC</A>] <BR><B>Antoine Guitton and Jon Claerbout</B> <BR><SMALL>In testing on several nearby data sets, three shown here,
<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep148/antoine3/paper_html/index.html">Six tests of sparse log decon</A> [<A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep148/antoine3/paper.pdf"><img height=15 src=icons/pdf.png>PDF 2.8M</A> <A HREF="http://sepwww.stanford.edu/data/media/private/docs/sep148/antoine3/antoine3.tar.gz"><img height=15 src=icons/GZIcon.png>SRC</A>] <BR><B>Antoine Guitton and Jon Claerbout</B> <BR><SMALL>Previously, we developed a sparseness-goaled decon method.
 HREF="antoine3/paper_html/node7.html#claerbout.sep.148.jon">Claerbout and Guitton, 2012</A>).