基于局部相似属性的初至自动拾取

Automatic first break picking based on local similarity attribute

本文提出了一种基于局部相似属性的初至自动拾取方法。首先是由炮记录中的地震道计算其局部互相关相似属性,形成局部相似属性道;然后根据粗略估计的近地表速度和炮检距对炮集中的每一道计算一个初至搜索中心,并以此中心确定每一道的搜索范围;最后将这些局部属性道作为能量比法的输入,计算局部相似属性道的能量比曲线,取搜索范围内能量比曲线的最大值所在位置为初至时刻。模型数据和实际资料试验结果表明,该方法具有一定的抗噪性能,且易于使用。

We propose in this paper a new approach for automatic first break picking.First,local similarity attributes such as local cross-correlation are calculated on traces of shot records.At the same time,a central point of first-break search is calculated based on near-surface velocity and offset for every trace of shot gathers.And a first-break search scope is determined for every trace of the shot gather.Then local cross-correlation trace is used as the input of the modified energy ratio method to detect first-breaks.The location of the maximum energy ratio in the search range can be considered as first break of the trace.Model and real data tests show that the proposed approach is robust under noisy conditions and easy to use.