Simultaneous estimation of local intersecting slopes

Abstract

The local slope of a seismic event is one of the most important attributes in seismic data processing and interpretation. In simple cases, only one slope value corresponds to a data location; however, in some complex geologic environments, such as unconformities or faults, two or more seismic events may intersect and yield several slopes. Although there are well-established methods for computing slopes in simple cases, there is no satisfactory algorithm in complex situations. We first detail the multimodality of coherence of intersecting events in the slope domain, and then propose a novel multimodal optimization method to capture their local slopes simultaneously. In the method, we use the neighborhood-based crowding differential evolution (NCDE) multimodal optimization algorithm cascaded by the hill-valley niche detection function and the Nelder-Mead simplex local search algorithm. Using this approach, we can obtain all of the local intersecting slopes simultaneously and avoid the need for a fine search when fully adopting the NCDE algorithm. The results of our numerical analysis indicate that the proposed method is robust, efficient, and accurate. Furthermore, the application of the proposed method to the dip-steering median filter for random noise attenuation shows better results than the iterative [Formula: see text] method in the case of complex wavefields.