At present, most signal-to-noise ratio (SNR) estimation methods can only calculate the global and not the local SNR of seismic data. This paper proposes a calculation method of a structure-oriented-based seismic SNR attribute. The purpose is to characterize the temporal and spatial variation of the seismic data SNR. First, the local slope parameters of the seismic events are calculated using a plane wave decomposition filter. Then, the singular value decomposition method is used to calculate the local seismic data SNR, thereby obtaining it in time and space. The proposed method overcomes the insufficiency of a conventional global SNR to characterize any local seismic data features and uses the SNR as an attribute of seismic data to more accurately describe the signal-noise energy distribution characteristics of seismic data in time and space. The results of a theoretical model test and real data processing show that the SNR attribute can be used not only for seismic data quality evaluation but also for analysis and evaluation of denoising methods.

中文翻译：

---

基于结构取向的地震数据信噪比属性计算方法

目前，大多数信噪比（SNR）估计方法只能计算地震数据的整体SNR，而不能计算局部SNR。提出了一种基于结构的地震信噪比属性的计算方法。目的是表征地震数据SNR的时空变化。首先，使用平面波分解滤波器计算地震事件的局部斜率参数。然后，使用奇异值分解方法计算局部地震数据SNR，从而在时间和空间上获得它。所提出的方法克服了传统的全局SNR不足以表征任何局部地震数据特征的问题，并使用SNR作为地震数据的属性来更准确地描述地震数据在时间和空间上的信噪能量分布特征。