The paper 🔺Xintao Chai et al., 2020, Deep learning for irregularly and regularly missing data reconstruction🔺 uses mixed synthetic and field datasets to train U-net like the neural network to combat the problem of missing data.
Once the network is fully trained, they test it on a diverse field data set and compare it with the Fourier transform interpolation method. The testing shows the effectiveness, superiority, and generalization capability of the proposed technique.
Link to the paper: https://www.nature.com/articles/s41598-020-59801-x
Link to the LinkedIn post