【信息技术】【2018.12】离散频率地震学

本文为荷兰代尔夫特理工大学（作者：Adriaan Visser）的硕士论文，共62页。

这项工作决定了数据中存在的频率分量的数量是否可以减少，同时仍然能够保持图像质量，而地震研究中的大多数工作都是在减少空间采样方面进行的。用主成分分析法对若干数据集的频谱分析表明，陆上地震数据中确实存在大量的频率冗余，并试图按重要程度生成频率分布。

考虑到陆上地震数据频谱的这种冗余性，人们尝试通过迭代应用傅里叶变换来重建缺失的频率。然而，该变换没有考虑空间采样，其目的是补偿丢失的频率。因此，我们选择使用线性Radon变换来代替变换域中的时空连通分量。建立了一个CGNE方案来重建数据，该方案在数据记录中的几乎线性渐近线上表现良好，频率分量减少了70%。该方案迭代地将线性Radon变换应用于记录数据，用基于范数的幅度对变换域中的数据进行加权。由于频率丢失而产生混叠而分散的能量被重新聚焦到主反射器上，特别是沿着反射双曲线的渐近线。缺失的频率被重建后达到一个比例因子，能很好地填充高达6Hz的带隙。

接下来，本文试图给出定量的质量指标，使地震图像之间的比较更容易，且是基于数据出发的，而不是主观的视觉判断。将偏移视为一个黑匣子，为偏移的部分设计了几种质量度量指标：与真值的相关性、图像内的对比度、发现的直线平均长度和局部信噪比。对比度不是比较图像的一个很好的指标，因为它在图像上的平均值几乎是恒定的。其他参数都是很好的度量，并且显示出一个明显的趋势，即数据记录中出现的频率分量越少，最终图像的质量就越差。观察到图像质量恶化的增加大约减少了70%，这与先前发现的数据记录值一致。

This work determines whether the amount of frequency components presentin the data can be reduced, whilst still retaining image quality, whereas mostefforts in seismological research are done in reducing spatial sampling. It isshown using a PCA on the frequency spectra of several data sets that indeed a largeredundancy in frequency content is present in onshore seismic data, and anattempt is made to generate a distribution of frequencies in order ofimportance.
Given this redundancy in the frequency spectrum ofonshore seismic data, it has been attempted to reconstruct the missingfrequencies by applying the Fourier transformation iteratively to the data. However,this transform does not take spatial sampling into account, which is aimed atto compensate for the missing frequencies. Therefore it has been elected to usea linear Radon transformation instead, which keeps components which areconnected in space-time connected in the transform domain. A CGNE scheme hasbeen set up to reconstruct the data, which performs very well along the almost linearasymptots in the shot records, up to a reduction of 70% of frequencycomponents. This scheme iteratively applies the linear Radon transform to ashot record, weighing the data in the transform domain with an amplitude basednorm. The energy that was spread out due to aliasing because of the missingfrequencies is refocused to the main reflectors, especially along the asymptotsof the reflection hyperbola. Missing frequencies are reconstructed, up to a scalingfactor, and band gaps of up to 6Hz get filled in very well.
Next, it is attempted in this work to give quantitativequality metrics, to make comparison between seismic images easier and based ondata, rather than subjective visual inspection. Treating migration as a blackbox, several quality metrics have been devised for the migrated sections:correlation to the ground truth, contrast within an image, average length offound lines, and local SNR. Contrast is not a very good metric to comparebetween images as its average across an image is almost constant with reductionpercentage. The other parameters are good metrics and show a clear trend thatthe fewer frequency components present in the shot records, the worse thequality of the final image. An increase in deterioration of image quality isobserved around 70% reduction, which is in correspondence with the earlierfound value for the shot records.

1. 引言
2. 频谱中冗余度的确定
3. 空间样本缺失的傅里叶重建
4. 频率缺失的Radon重建
5. 质量描述与结果
6. 结论、讨论与建议
   附录 其它表格与绘图

更多精彩文章请关注公众号：