基于高精度曲率分析的断层识别方法

Fault Identification Based on High-precision Curvature Analysis

为了克服常规相干体分析抑制噪声能力较差及对小断层识别能力较差的缺陷,基于自适应中值滤波和曲率分析,提出了一种新的基于高精度曲率分析的断层识别方法。该方法首先进行自适应中值滤波去噪,在滤除随机噪声的同时较好地保持了图像的细节;再进行最大正曲率计算;最后通过提取目的层位的最大正曲率切片进行断层解释。该方法充分利用了自适应中值滤波能较好去噪的优点,以及最大正曲率对断层高敏度性的优点。实际资料的计算结果表明,该方法简单易行,可以有效识别大、中、小等不同尺度的断层,断层识别率显著高于常规相干体方法。

In order to effectively overcome the defects of conventional coherence cube analysis that the function of suppressing noise and identifying small faults were not very effective.Based on adaptive median filter and curvature analysis a new high-precision curvature analysis was proposed based on the fault identification method.This method first removed noise by adaptive median filter,which could filter out random noise and maintain better image in detail for a maximum positive curvature calculation.At last,the fault is interpreted by extracting the maximum positive curvature slice of the horizon.The method takes full advantage of the better denoising effect of the adaptive median filter and the high sensitivity of the maximum positive curvature analysis for the fault.The actual data results show that the proposed high-precision curvature analysis is easy and efficient.