Efficient and accurate inversion of multiple scattering with deep learning

TL;DR

This paper introduces a deep learning approach for inverting multiple light scattering measurements, significantly improving speed and image quality over traditional optimization methods in diffraction tomography.

Contribution

It presents a novel deep convolutional neural network that directly inverts complex scattering data, replacing traditional nonconvex optimization techniques.

Findings

Faster image reconstruction compared to state-of-the-art methods

Higher quality images achieved with the deep learning model

Effective on both simulated and experimental datasets

Abstract

Image reconstruction under multiple light scattering is crucial in a number of applications such as diffraction tomography. The reconstruction problem is often formulated as a nonconvex optimization, where a nonlinear measurement model is used to account for multiple scattering and regularization is used to enforce prior constraints on the object. In this paper, we propose a powerful alternative to this optimization-based view of image reconstruction by designing and training a deep convolutional neural network that can invert multiple scattered measurements to produce a high-quality image of the refractive index. Our results on both simulated and experimental datasets show that the proposed approach is substantially faster and achieves higher imaging quality compared to the state-of-the-art methods based on optimization.