Transfer Learning Enhanced Full Waveform Inversion
arXiv:2302.11259 · doi:10.1109/AIM46323.2023.10196158
Abstract
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.
arXiv comment7 pages, 5 figures
References in corpus (10)
- Adam: A Method for Stochastic Optimization
- Scikit-learn: Machine Learning in Python
- Artificial Neural Networks for Solving Ordinary and Partial Differential Equations
- Visualizing the Loss Landscape of Neural Nets
- Physics-informed Neural Networks (PINNs) for Wave Propagation and Full Waveform Inversions
- Deep-Learning Inversion of Seismic Data
- On weight initialization in deep neural networks
- On the Use of Neural Networks for Full Waveform Inversion
- The Neural Network Approach to Inverse Problems in Differential Equations
- Quantitative reconstruction of defects in multi-layered bonded composites using fully convolutional network-based ultrasonic inversion