Improved image segmentation for tracking salt boundaries

Jesse Lomask, Biondo Biondi and Jeff Shragge

lomask@sep.stanford.edu, biondo@sep.stanford.edu, jeff@sep.stanford.edu

ABSTRACT

Normalized cut image segmentation can be used to track salt boundaries within an iterative velocity analysis scheme. To overcome the formidable computational expense and storage requirements of normalized cut image segmentation, three cost saving approaches are proposed. First, pixels are sampled from windows centered at powers of 2. This greatly increases the sparseness of the weight matrix. Second, initial solutions are provided to subsequent segmentations for multiple segmentation passes in iterative velocity analysis. Third, an iterative multi-scale approach would be necessary for tracking of the bright salt events in large 3D cubes.