Seismic anisotropy is present in different geological settings at various scales. This significantly affects the standard hyperbolic reflection moveout approximation that is based on the assumption that layers are isotropic. Conventional traveltime picking and velocity analysis picking can take up a significant amount of time to process anisotropic seismic data. Here we develop a velocity-independent Hough transform neural network based technique for normal moveout correction in VTI media. This technique offers advantages when compared to the time and computational costs required in conventional anisotropic normal moveout correction. Using a Hough transform based neural network, we simultaneously fit all the non-hyperbolic reflection moveout curves using intermediate to long offsets. The functionality of the unsupervised neural network is analogous to the classical Hough transform. However, the presented form is an efficient representation with reduced storage space and is capable of handling noisy datasets. The network is applied to two synthetic VTI datasets and we demonstrate the practical feasibility of anisotropic moveout correction that is independent of traveltime picking and velocity analysis
Presentation Date: Wednesday, September 18, 2019
Session Start Time: 8:30 AM
Presentation Time: 11:00 AM
Location: 221D
Presentation Type: Oral
