A fault surface extraction and reconstruction method based on 3D seismic image

Abstract

Summary Fault interpretation, a main part of the interpretation of 3D seismic images, is an important means to the exploration of complex geological structures. Conventional fault interpretation methods are manual, which are difficult, time consuming and subjective. Recently, to improve the efficiency of interpretation, mass of surface auto-extraction methods have been presented. These methods mainly follow two steps. Firstly, data is processed to make fault prominent and then fault surface is reconstructed by the sticks which usually are generated from data slice. In this paper, we present a new flow of fault surface extraction and reconstruction which focuses on the use of 3D points cloud surface reconstruction idea to construct the fault plane. First of all, we build a new binary data in 3D space according to fault attribute body. Next, we use 3D image processing technology to preprocess the binary data. And then, we propose a fault surface reconstruction method which uses 3D spatial points cloud clustering based on fault surface features. At last, we use F3 data as example to extract fault surface. The experimental result meets our