Imaging of sub-wavelength micro-structures by time reversal and neural networks, from synthetic to laboratory-controlled data
Abstract
Imaging of a sub-wavelength micro-structure com- posed of a periodic grid-like finite set of circular rods is carried out from transient scattered field data in different configurations of sources and receivers. The goal is to identify the position of possibly missing rods. Time reversal is confirmed as a cheap yet efficient first-order diagnostic method even in the demanding con- text of a sub-wavelength micro-structure. Tools of deep learning, expected to be valid in very general circumstances if data can be acquired in sufficient number, are in parallel developed to image the micro-structure. To that effect, recurrent neural networks (RNN) and convolutional neural networks (CNN) are both used. Pros and cons of all approaches are illustrated by comprehensive simulations from synthetic data computed via a FD-TD software carefully tailored to the micro-structure model, and used also to make the networks learn the micro-structures. The analysis is completed from examples on laboratory-controlled data acquired on a properly built micro-structure prototype set within a microwave anechoic chamber. These examples confirm the good promises of neural networks even with rather scarce data as exemplified in a forward scattering case —fixed source and receiver antennas face each other and the micro-structure is rotated between them.