3D Scattering Tomography by Deep Learning with Architecture Tailored to Cloud Fields

TL;DR

This paper introduces 3DeepCT, a deep learning model tailored for 3D scattering tomography of cloud fields, achieving faster and more accurate reconstructions than traditional physics-based methods, with a hybrid approach further enhancing performance.

Contribution

The paper presents a novel deep neural network architecture specifically designed for atmospheric cloud field scattering tomography, outperforming existing physics-based methods in speed and accuracy.

Findings

3DeepCT outperforms physics-based methods in accuracy

Significant orders of magnitude reduction in computation time

Hybrid model improves recovery performance

Abstract

We present 3DeepCT, a deep neural network for computed tomography, which performs 3D reconstruction of scattering volumes from multi-view images. Our architecture is dictated by the stationary nature of atmospheric cloud fields. The task of volumetric scattering tomography aims at recovering a volume from its 2D projections. This problem has been studied extensively, leading, to diverse inverse methods based on signal processing and physics models. However, such techniques are typically iterative, exhibiting high computational load and long convergence time. We show that 3DeepCT outperforms physics-based inverse scattering methods in term of accuracy as well as offering a significant orders of magnitude improvement in computational time. To further improve the recovery accuracy, we introduce a hybrid model that combines 3DeepCT and physics-based method. The resultant hybrid technique enjoys fast inference time and improved recovery performance.