Multiscale and multiaspect method of removal surface-wave based on the second generation Curvelet transform
Jingjing Zheng, Xingyao Yin, Guangzhi Zhang · 2009
In this paper, according to low-frequency , high-amplitude and low-apparent velocity characteristics of ground roll, we develop a new and effective multiscale and strong directional method of identifying and removing ground roll based on the second generation Curvelet transform. Making the best of multiscale and strong directional characteristics of the Curvelet transform, seismic data with noise is decomposed into different frequency band and different angle data. According to the different characteristics of surface waves on different frequency band and different angle data, we take different processing, and do not process the frequency band and angle data without surface waves. Applying this method to pre-stack seismic data, it can remove surface waves successfully, and at the same time protect effective wave of deep layer well, particularly, the low-frequency band efficiency data. This indicates the method in this paper is an effective and preserved-amplitude way.