基于时间窗互相关成像条件的GPR逆时偏移成像

Reverse time migration of ground penetrating radar based on time window cross-correlation imaging condition

针对基于互相关成像条件的探地雷达(GPR)逆时偏移在地下介质的非反射点处易产生强噪声的不足,根据偏移速度模型中每个成像点的主反射时间,设计一个时间窗函数以压制时间窗外的噪声干扰,提出了一种基于时间窗互相关成像条件的GPR逆时偏移算法。其中,时域有限差分法(FDTD)用于计算正传和反传电磁波场,时间窗内正传与反传电磁波场的归一化互相关计算结果作为权系数作用于基于归一化互相关成像条件的GPR逆时偏移剖面,拉普拉斯滤波用于压制低频噪声。在此基础上,编制了基于时间窗互相关成像条件的GPR逆时偏移程序,利用该程序对两个典型的GPR模型的正演数据和实测数据进行计算,结果表明:相比于归一化互相关成像条件的逆时偏移结果,基于时间窗互相关成像条件的GPR逆时偏移剖面的低频噪声更弱、杂波干扰更小、空间分辨率更高、成像精度更好。

As ground-penetrating radar (GPR) reverse time migration based on cross-correlation imaging condition, it is easy to generate strong noise in the reverse time migration profile of GPR at the non-reflection points of subsurface media. In this study, a time window function is designed and applied to the incident wave field and scattering wave field for each imaging point, to strengthen the correlation of the wave fields in the time domain. The finite difference time domain method (FDTD) is used to calculate the forward and backward electromagnetic wave fields. The normalized cross-correlation results of the forward and backward electromagnetic wave fields in the time window are used as the weight coefficients for normalized cross-correlation imaging. Laplacian filtering is used to suppress low-frequency noise. On this basis, a GPR inverse time migration program based on time window cross-correlation imaging condition was developed. Simulated data of two typical GPR models and observed data were calculated, and GPR inverse time migration was performed and compared with the normalized cross-correlation imaging condition. The results show that the GPR reverse-time migration imaging section based on the time-window cross-correlation imaging condition has low-frequency noise, less clutter, better spatial resolution and higher imaging accuracy, than that in the normalized cross-correlation imaging condition.