Seismic Instantaneous Frequency Extraction Based on the SST-MAW

Naihao Liu,Jinghuai Gao,Xiudi Jiang,Zhuosheng Zhang,Ping Wang
DOI: https://doi.org/10.1088/1742-2140/aa8cb6
2018-01-01
Journal of Geophysics and Engineering
Abstract:The instantaneous frequency (IF) extraction of seismic data has been widely applied to seismic exploration for decades, such as detecting seismic absorption and characterizing depositional thicknesses. Based on the complex-trace analysis, the Hilbert transform (HT) can extract the IF directly, which is a traditional method and susceptible to noise. In this paper, a robust approach based on the synchrosqueezing transform (SST) is proposed to extract the IF from seismic data. In this process, a novel analytical wavelet is developed and chosen as the basic wavelet, which is called the modified analytical wavelet (MAW) and comes from the three parameter wavelet. After transforming the seismic signal into a sparse time-frequency domain via the SST taking the MAW (SST-MAW), an adaptive threshold is introduced to improve the noise immunity and accuracy of the IF extraction in a noisy environment. Note that the SST-MAW reconstructs a complex trace to extract seismic IF. To demonstrate the effectiveness of the proposed method, we apply the SST-MAW to synthetic data and field seismic data. Numerical experiments suggest that the proposed procedure yields the higher resolution and the better anti-noise performance compared to the conventional IF extraction methods based on the HT method and continuous wavelet transform. Moreover, geological features (such as the channels) are well characterized, which is insightful for further oil/gas reservoir identification.
What problem does this paper attempt to address?