Optimum Image from Sub-surface by Pre-stack Multiples suppression; Parabolic Radon Transform Approach

Multiples suppression is a partially solved problem and there is not a precise solution for it. Although stacking is a good tool for multiples suppression but some times we should suppress them before stacking. Generally pre-stack multiples suppression methods work on the basis of move out difference between primaries and multiples. Many authors investigated methods for filtering of this type of coherent noises using this characteristic. Radon transform is one of the most frequent used methods in this subject and there are many papers on it. Radon transform has been classified due to its applications and formulations: linear, hyperbolic and parabolic. The first one is suitable for linear events classification and elimination like ground roles; second type is ideal for hyperbolic events classification and suppression but unfortunately is time consuming, because one encounters with nonlinear equations that are difficult to formulate on the computer. Parabolic version was proposed by Hampson (1986); parabolic curves display curvature but does not need taking a square root and handling difficult nonlinear equations. After Radon transform, in tau-p panel we can filter undesired events. In this paper we programmed this algorithm and showed its efficiency in tackling with strong multiples both in synthetic and real data sets.