基于曲波变换与联合方法的地震数据重建

Seismic Data Reconstruction Based on Curvelet-transform and Joint Method

野外地震数据采集通常会受地形等不可控因素的影响,导致采集到不规则缺失的地震数据,为了不影响后续地震资料处理,需要对缺失的地震数据进行重建。传统的重建算法往往收敛速度较慢,为此,提出将快速阈值迭代法和线性Bregman方法进行联合,在快速阈值迭代加速的基础上进一步提高了算法的效率,得到了一种快速高精度重建方法。在重建的过程中,以曲波域作为稀疏变换域,采用硬阈值函数和指数阈值模型。通过理论数据的模拟和实际数据的检验表明,相较于传统的地震数据重建方法,该方法重建的效果明显,重建效率更高。此外,该方法具有一定的抗噪性,在快速阈值迭代法高效收敛的基础上进一步加快了重建的效率。

The collection of seismic data in field surveys is often affected by uncontrollable factors such as terrain,resulting in irregular and missing seismic data.In order to not compromise the subsequent seismic data processing,it is necessary to reconstruct the missing seismic data.However,traditional reconstruction algorithms often have slow convergence rates.To address this issue,a new joint operator is proposed,which combines the fast threshold iterative method with the linear Bregman method.This combination further improves the efficiency of the algorithm on the basis of the acceleration of fast threshold iteration method,resulting in a fast and high-precision reconstruction method.During the reconstruction process,the curvelet domain is used as the sparse transform domain,and the hard thresholding function and exponential threshold model are employed.Theoretical data simulations and practical data verification demonstrate that compared to traditional seismic data reconstruction methods,this method achieves significant improvements in reconstruction quality and efficiency.Additionally,this method exhibits certain noise robustness.By leveraging the efficient convergence of the fast threshold iterative method,the reconstruction speed is further accelerated.