We introduce an analytical method for integrating dip information to flatten uninterpreted seismic data. First, dips are calculated over the entire seismic volume. The dip is then integrated in the Fourier domain, returning for each sample a time shift to a flat datum. Then each sample is shifted in the seismic data to remove all structural folding deformation in a single non-interpretive step. Using the Fourier domain makes it a quick process but requires that the boundaries are periodic. This method does not yet properly handle faults because of their discontinuous nature, but is presently very effective at removing warping and folding.

Delineating salt boundaries is a necessary step in the velocity-model building process. The salt-delineation problem can be thought of as an image-segmentation problem. Normalized cuts image segmentation (NCIS) finds the cut (or cuts) that result in an image being broken into portions which have dissimilar, by some measure, characteristics. We apply a modified version of the NCIS method to partition seismic images along salt boundaries. NCIS can track boundaries that are not continuous, where conventional horizon-tracking algorithms may fail, by calculating a weight connecting each pixel in the image to every other pixel within a local neighborhood. The weights are determined using problem-dependent combinations of attributes, the most important being instantanteous amplitude and dip. The weights for the entire image are used to segment the image via an eigenvector calculation. The weight matrices for 3D seismic data cubes can be quite large and computationally expensive. By imposing bounds and by distributing the algorithm on a parallel cluster, we significantly increase efficiency. This method is demonstrated to be effective on a 3D field seismic data cube.

SUMMARYImage segmentation can be used to track salt boundaries when the salt boundary amplitude is greater than any other local reflections. We apply a modified version of the normalized cut image segmentation method to partition seismic images along salt boundaries. In principle our method should work even when the boundaries are not continuous, and conventional horizon tracking algorithms may fail. Our implementation of this method calculates a weight connecting each pixel in the image to each pixel in a local neighborhood. The weight is made weak where the negative amplitude of the complex trace along the shortest path between the two pixels has a minimum and is less than a threshold value. This method is demonstrated to be effective on synthetic 2D seismic sections and can easily be modified to be applied to 3D data. To overcome the formidable computational expense and storage requirements, three cost saving approaches are proposed. Firstly, pixels are sampled from windows centered at powers of 2, this greatly increases the sparseness of the weight matrix. Secondly, initial solutions are provided to subsequent segmentations for multiple segmentation passes in iterative velocity analysis. Thirdly, an iterative multiscale approach should allow the tracking of the bright salt events in large 3D cubes.

Volumetric flattening is a method for automatically flattening entire 3D seismic cubes without manual picking. This is an efficient algorithm that intrinsically performs automatic dense picking on entire 3D cubes at once. The method involves first calculating local dips (step-outs) everywhere in the data using a dip estimation technique. These dips are resolved into time shifts (or depth shifts) via a non-linear least-squares problem. The data are subsequently vertically shifted according to the summed time shifts to output a flattened volume. A most attractive feature of the previously described flattening technique is that it requires no manual picking. This would be fine if all data sets had reasonably accurate estimated dips, but in the real world automatic flattening can produce results that are not perfect. Noise, both coherent and otherwise, can overwhelm the dip estimation causing reflectors in those areas to not be flat. Multiples, for example, are significant source of coherent noise that can contaminate the dip field. Although certain faults with tip-lines (terminations) encased within the data cube can be flattened if a fault model is provided, faults that cut across the entire data cube cannot. Consequently, it would be useful to have the ability to add some geological constraints to restrict the flattening result in areas of poor data quality while allowing it to efficiently tackle other areas where estimated dips are more