Machine learning based data retrieval for inverse scattering problems with incomplete data
arXiv:1910.12745
Abstract
We are concerned with the inverse scattering problems associated with incomplete measurement data. It is a challenging topic of increasing importance in many practical applications. Based on a prototypical working model, we propose a machine learning based inverse scattering scheme, which integrates a CNN (convolution neural network) for the data retrieval. The proposed method can effectively cope with the reconstruction under limited-aperture and/or phaseless far-field data. Numerical experiments verify the promising features of our new scheme.
arXiv commentThe authors withdrew the previous three versions because more work was needed to furnish before its appearance as a formal paper. The current version is fine. All helpers agree with current version
References in corpus (7)
- Adam: A Method for Stochastic Optimization
- Provable approximation properties for deep neural networks
- Extrapolated full waveform inversion with deep learning
- Stable determination of sound-hard polyhedral scatterers by a minimal number of scattering measurements
- Retrieval of acoustic sources from multi-frequency phaseless data
- Recovering an electromagnetic obstacle by a few phaseless backscattering measurements
- Data recovery: from limited-aperture to full-aperture