基于多窗口相干性的倾角导向主分量滤波

为了在压制地震数据噪声的同时保持反射结构边缘信息和有效频率成分，研制了基于多窗口相干性的倾角导向主分量滤波方法，该法结合边缘保持滤波，可实现多窗口边界保持和主分量滤波。将此法用于多道地震记录提取有效信号的特征值重构信号，可以提高地震记录的信噪比。利用多窗口相干性扫描和二次曲线/面拟合计算二维/三维地层的精确视倾角，以相干值最大时的倾角导向窗口内的地震道数据作为主分量滤波的输入，实现输入主分量滤波器的子体数据的自动选择。实际资料的应用和分析表明，基于多窗口相干性的倾角导向主分量滤波方法在增强同相轴连续性的同时，也加强了有用的不连续性信息（如断层），避免了由于滤波作用而引起的地震图像边缘模糊，可清晰地刻画地层的几何形态和展布特征，且能够较好地保持原始地震数据体的有效频率成分和地震纵向分辨率。

Dip-steering principal component filter using multi-window coherence scanning and interpolation

We propose in the paper a dip-steering principal component filter based on multi-window coherence scanning and interpolation in order to remove seismic noise and simultaneously preserve reflection structure boundary and effective frequency components. Combining edge-preserving filtering, this method implements edge preservation and principal component filtering in multi-windows. Multi-trace eigenvalues are extracted from seismic records and seismic signals are reconstructed to improve data S/N ratio. 2D/3D apparent dips are calculated by multi-window coherence scanning and quadratic curve/surface fitting. Seismic data in the dip steering window with the maximum coherence value is aquatically selected as an input of the principal component filter. Real data applications show that the proposed method can preserve useful components and bandwidth of seismic data, and improve seismic vertical resolution. Therefore it voids fuzzy edge caused by filtering, and depicts clearly formation spatial distribution.