We propose a way to favorably employ neural networks in the field of non-destructive testing using Full Waveform Inversion (FWI). The presented methodology discretizes the unknown material distribution in the domain with a neural network within an adjoint optimization. To further increase efficiency of the FWI, pretrained neural networks are used to provide a good starting point for the inversion. This reduces the number of iterations in the Full Waveform Inversion for specific, yet generalizable settings.

翻译：我们提出了一种在全波形反演（FWI）领域有利地使用神经网络进行无损检测的方法。所提出的方法在伴随优化过程中，利用神经网络对未知材料分布进行区域离散化。为进一步提升FWI的效率，采用预训练神经网络为反演提供良好的初始点。这减少了在特定且可泛化的场景下全波形反演所需的迭代次数。