arXiv Machine Learning
Multi-Level-Set-Based Physics-Driven Neural Network to Solve 3-D Inverse Scattering Problems
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv Machine Learning
Oct 15
Structured Multi-Level-Set Inversion for 3-D Electromagnetic Inverse Scattering
The paper introduces Structured Multi-Level-Set Inversion (SMLSI), a physics‑driven neural network approach for 3‑D electromagnetic inverse scattering that represents contrast using object support and material distribution rather than direct parameterization. SMLSI employs multiple level‑sets mapped to soft‑support functions, a soft‑union operation, and trainable complex permittivities, with structure‑aware regularization to enforce binary support and coherent material regions. Numerical and experimental results show that SMLSI outperforms conventional methods like CSI and $L_{2/3}$‑FBE‑WCIE in reconstructing irregular geometries, closely spaced objects, and multiple material regions from scattered field data.
arXiv Machine Learning
Aug 11
Coordinate-Residual Physics-Driven Neural Network for Electromagnetic Inverse Scattering
arXiv:2608. 09382v1 Announce Type: cross Abstract: Electromagnetic inverse scattering is a nonlinear and ill-posed problem, where accurate reconstruction is challenging due to measurement limitations, noise, and high computational costs, especially for 3-D imaging.
arXiv Machine Learning
Sep 2
Coordinate-Residual Physics-Driven Neural Network for Inverse Scattering Imaging
The paper introduces a coordinate-residual physics-driven neural network (CRPDNN) for 3‑D electromagnetic inverse scattering. CRPDNN models the unknown contrast distribution using normalized spatial coordinates and a residual convolutional network, optimizing parameters by enforcing consistency between measured and predicted scattered fields. It eliminates the need for preliminary reconstruction, achieving lower relative error and significant speedups compared to existing methods, while maintaining stability under noisy measurements and showing promise in practical imaging experiments.
arXiv Machine Learning
Aug 5
ScoreField: Neural Inverse Scattering with Score-Based Generative Priors
arXiv:2608. 02937v1 Announce Type: cross Abstract: Designing an effective electromagnetic inverse-scattering solver requires faithful enforcement of nonlinear full-wave physics together with an expressive prior on the unknown permittivity contrast.
arXiv Machine Learning
Jun 29
Recovering Sharp Conductivity Features in the Finite-Data Calder\'on Problem with Physics-Informed Neural Networks
arXiv:2606. 28158v1 Announce Type: new Abstract: Physics-informed neural networks (PINNs) have recently emerged as a promising framework for addressing the Calder\'on inverse problem from limited boundary data.
arXiv AI
Jul 28
AI-interpreted Optical Scattering for Robust and Focal Depth-Aware Imaging
arXiv:2607. 22867v1 Announce Type: cross Abstract: Optical scattering has conventionally been regarded as an impediment in imaging research due to the degradation of image quality during reconstruction.