摘要
大多数的地震数据去噪方法都是根据单参数特征信息提取有效信号或噪声，达到去噪目的。本文利用神经元网络的高维非线性映射特征，设计一种非线性滤波器对数据进行非线性变换。这种滤波器与常规滤波器的不同点在于它综合考虑了地震有效信号和噪声的多种参数表现特征，并将两者有效地分离。经这种方法去噪（空间域）的数据剖面不会降低分辨率，去噪效果受数据信噪比和反射波同相轴倾角的影响较小。本文展示了空间域的处理结果，该方法也适用于其它域。
Most methods for seismic noise removal use the characteristic information ofsingle parameter to separate out useful signals or noises, thus achieving usual noiseremoval. A nonlinear filter for nonlinear transform of seismic data can be designedby taking high dimensional nonlinear mapping character of neural network. Such afilter, differing from other common ones, comprehensively analyses many differ-ences between signal and noise to separate off them effectively. With the use of thisnoise removal method (in space domain), the seismic section retains its original res-olution, and noise removal effect suffers little damage which results from low sig-nal/noise ratio and the dip angle of reflection leg. The paper only displays the pro-cessing results in space domain. This noise removal method is also applicable toother domains.
出处
《石油地球物理勘探》
EI
CSCD
北大核心
1998年第1期104-108,共5页
Oil Geophysical Prospecting
关键词
滤波
去噪
非线性
地震数据
地震勘探
filtering, nonlinear trans formation, neurone, network, pararneter, characteristic, noise elimination
