AN IMPROVED PROJECTION ONTO CONVEX SETS METHOD FOR SIMULTANEOUS INTERPOLATION AND DENOISING

Jingjie Cao, Benfeng WangChinese 2015 · DOI

Abstract

Abstract Sparsity‐promoted seismic interpolation is an important interpolation strategy; however, most existing interpolation methods merely focused on high signal‐to‐noise ratio data. Actually, noise is always contained in sampled data; the existence of noise will deteriorate interpolation results and makes interpolation challenging. Projection onto convex sets (POCS) method is an efficient method; however, it cannot get acceptable results for noisy data interpolation. A strategy for simultaneous interpolation and denoising is the weighted POCS method which uses an additional weighting factor to achieve denoising. In this paper, we demonstrate that the weighted projection onto convex sets (POCS) method is a method for unconstrained optimization model, and it can be deduced from the iterative thresholding method. In addition, an improved POCS method is proposed for interpolation and denoising without extra computation effort. Numerical examples on synthetic and field data demonstrate that the improved POCS method can get reliable results for noiseless and noisy data interpolation. First interpolation and then denoising examples proved that simultaneous interpolation and denoising is robust when the least thresholds are properly chosen.
BibTeX
@inproceedings{cao2015improved,
  author = {Jingjie Cao and Benfeng Wang},
  title = {{AN IMPROVED PROJECTION ONTO CONVEX SETS METHOD FOR SIMULTANEOUS INTERPOLATION AND DENOISING}},
  booktitle = {Chinese},
  year = {2015},
  doi = {10.1002/cjg2.20195},
}

Loading related content...

Figure