论文部分内容阅读
在地震勘探数据采集中,随机噪声严重影响地震资料质量,给后期解释工作带来很大困难。如何在不损失剖面有效信息的前提下压制随机噪声,有效地提高地震资料的信噪比和保真度,是本文的研究目标。构造导向滤波技术的核心是构造方向表征的求取以及如何实现非平稳滤波,来达到提高地震数据信噪比和保真度的目的。本文首先通过分析函数二维导数与希尔伯特变换的频率响应关系,推导出了基于二维希尔伯特变换的非迭代地震同相轴倾角求取算子,进而达到了构造方向表征的求取;其次选取多项式拟合作为构造导向滤波中的非平稳滤波方法,扩展了非平稳多项式拟合的应用范围;最后沿构造倾角方向进行变振幅同相轴的非平稳多项式拟合,实现和构建了新的自适应构造导向滤波方法。理论模型和实际地震资料处理的结果表明,所提出的方法实现了既保护构造信息又有效地压制了随机噪声的目的。
In seismic data acquisition, random noise seriously affects the quality of seismic data, which brings great difficulties to the later interpretation. How to suppress the random noise under the premise of not losing effective information of the profile and effectively improve the signal-noise ratio and fidelity of seismic data is the research goal of this paper. The core of structure-oriented filtering technology is to find the direction of structural features and how to achieve non-stationary filtering to achieve the purpose of improving signal-to-noise ratio and fidelity of seismic data. In this paper, firstly, the frequency response of two-dimensional derivative of function and Hilbert transform is analyzed, and the dip-angle operator of non-iterative seismic events based on two-dimensional Hilbert transform is deduced, and then the structure orientation characterization Secondly, the polynomial fitting is chosen as the non-stationary filtering method in the structure-oriented filtering, which expands the application scope of the non-stationary polynomial fitting. Finally, the non-stationary polynomial fitting of variable amplitude events along the structural dip angle is implemented and constructed A new adaptive structure-oriented filtering method is proposed. The results of theoretical model and actual seismic data processing show that the proposed method realizes the purpose of both protecting the structure information and effectively suppressing the random noise.