We propose an efficient iterative curvelet-regularized deconvolution algorithm that exploits continuity along reflectors in seismic images. Curvelets are a new multiscale transform that provides sparse representations for images (such as seismic images) that comprise smooth objects separated by piece-wise smooth discontinuities. Our technique combines conjugate gradient-based convolution operator inversion with noise regularization that is performed using non-linear curvelet coefficient shrinkage (thresholding). The shrinkage operation leverages the sparsity of curvelets representations. Simulations demonstrate that our algorithm provides improved resolution compared to the traditional Wiener-based deconvolution approach.
| Identifer | oai:union.ndltd.org:LACETR/oai:collectionscanada.gc.ca:BVAU.2429/503 |
| Date | January 2005 |
| Creators | Hennenfent, Gilles, Herrmann, Felix J., Neelamani, Ramesh |
| Publisher | European Association of Geoscientists & Engineers |
| Source Sets | Library and Archives Canada ETDs Repository / Centre d'archives des thèses électroniques de Bibliothèque et Archives Canada |
| Language | English |
| Detected Language | English |
| Type | text |
| Rights | Herrmann, Felix J. |
Page generated in 0.0023 seconds