数学形态学在地震资料处理中的应用探索

Application of mathematical morphology in seismic data processing

传统的滤波方法一般是根据地震波的频率、波数、时差、振幅的差异进行去噪处理 ,如 f k滤波 ,f x 滤波、带通滤波、K L变换、拉冬变换等。对于山地地震资料 ,这些滤波方法的应用受到限制。可以从波形形态的细微差异上寻求新的滤波方法。根据数学形态学的滤波性质 ,考虑将其用在地震资料处理中。主要讨论了结构元素的尺寸、形状和幅值对处理结果的影响 ,并对实际地震资料做了处理。数学形态学的开闭运算 ,能滤除比结构元素小的亮细节和暗细节 ,可用其去除地震资料中的高频噪者。

Traditionall filtering methods, such as f-kfiltering,f-x filtering, bandpass filtering,K-L transform and Radon transform, eliminate the noises in seismic waves by the difference in frequency, wave number, time difference and amplitude. However, for seismic data acquired from mountainous regions, these methods fail to yield satisfactory results. In this paper, we try to eliminate the noises by the fine difference of wave shapes. Methods from mathematical morphology were adopted in seismic data processing.We studied the effects of the size and shape of structural elements on signals. Results from real data processing were presented.