【摘要】 传统的地震资料拓频方法大多采用反褶积、反Q滤波等方法，对中高频进行补偿以拓宽地震频带，但上述方法在低频拓展方面效果不太理想。在压缩感知框架下，利用反射系数反演研究了在频带双向拓宽的基础上提高地震资料分辨率的方法。首先根据压缩感知理论，利用反射系数的稀疏性和地震资料的可压缩性，构造了稀疏规则算子约束下的目标函数；然后通过快速软阈值迭代法反演反射系数，利用反射系数频谱中的低频及高频成分补偿地震资料的频谱，实现地震资料频谱的重构；最终达到频带双向展宽以提高地震资料分辨率的目的。模型及实际地震资料测试结果表明，利用上述方法可同时对低频端和高频端进行双向拓宽，在不改变原始地震剖面特征的同时，可以使地震资料有效频带拓宽20 Hz以上，并且提高了地震资料识别薄储层的能力，为“两宽一高”地震资料处理提供了技术支撑。更多还原

【Abstract】 Traditional frequency extension methods for seismic data use deconvolution, anti-Q filtering, etc.Although they can broaden the seismic frequency band by a certain extent by compensating the energy of medium and high frequencies, the effect of low-frequency expansion is not ideal.In this study, based on reflection coefficient inversion, the method of bidirectional frequency band expansion of seismic data was studied in the framework of compressed sensing.Based on compressed sensing theory, the sparsity of the reflection coefficient and compressibility of seismic data were used to construct the objective function under the constraint of the sparse rule operator.The reflection coefficient was inverted by the fast soft threshold iteration method, and the spectrum of the seismic data was reconstructed using the spectrum of the reflection coefficient to bi-directionally broaden the seismic data frequency band and improve the resolution.The test results of model and actual data showed that this method can widen the low and high frequencies in both directions.The effective frequency band of the seismic data was widened by more than 20 Hz without changing the characteristics of the original seismic profile, thereby improving the resolution of seismic data to identify thin reservoirs and providing technical support for seismic processing with a wide azimuth, wide frequency band, and high density.更多还原