Abstract:Electrical impedance tomography (EIT) is a promising functional and structural imaging method in process tomography. However, due to the "soft-field" nature and the high dependence on the prior information, it often suffers serious artifacts in quantitative analysis. Most recently, EIT image quality has significantly improved because of the state-of-the-art (SOTA) deep learning-based models in the aspect of solving the inverse problem, especially fully convolutional networks (FCNs) and V-Net variants. Despite their success, these deep convolutional neural networks (CNNs) have two limitations: 1) the long-range information transition is frequently lost and the reverse gradient often disappears in deep CNNs and 2) some novel skip connections, such as residual and dense connections, often occupy substantial computational resources. To overcome these two limitations, we propose -Net, a new neural architecture based on redesigned feature transited connections by the terms of: 1) a prereconstructor based on the iterative Newton–Raphson (NR) method, which maps the nonlinear function between the measurements and the initial images; 2) dual cascaded V-Nets are combined, which play the role of an encoder and a decoder, respectively; 3) a new parallel attention mechanism via channel attention and coordinate attention to learn the conductivity distributions and boundary-shaped feature separately; and 4) the lightweight skip connections reduce the computational resources (or accelerate the inference speed) of EIT imaging. The -Net is evaluated by using the multiphase flow industrial applications, and the results demonstrate that: 1) -Net has a better performance in shape reconstruction with sharp "corner"; 2) -Net could reconstruct the model accurately where it has some low-contrast conductivity distributions; 3) -Net enhances the quality of interfaces with the stratified flow; and 4) the pruned -Net achieves significant speedup compared with the VDD-Net or -Net. The analyses show that the average relative error is 0.05, the average correlation coefficient (CC) is 0.92, and the average structural similarity (SSIM) is 0.92 on the testing datasets. In addition, the average relative cover ratio is 0.97, and the average relative contrast (RCo) ratio is 0.98 on the testing datasets.

Electron density profile reconstruction with Convolutional Neural Networks

Plasma electron density profile tomography for EAST based on integrated data analysis

Prediction of plasma rotation velocity and ion temperature profiles in EAST Tokamak using artificial neural network models

Combined Analysis of Laser Interferometer and Microwave Reflectometer for a Consistent Electron Density Profile on HL-2A

Reconstruction of Poloidal Magnetic Fluxes on EAST based on Neural Networks with Measured Signals

Prediction of Ionospheric Electron Density Distribution Based on CNN-LSTM Model

Real-time equilibrium reconstruction by neural network based on HL-3 tokamak

Fast Estimation of Ion Temperature from EAST Charge Exchange Recombination Spectroscopy Using Neural Network

Neural networks for reconstruction and uncertainty quantification of fast-ion phase-space distributions using FILD and INPA measurements

Real-time equilibrium reconstruction by multi-task learning neural network based on HL-3 tokamak

Effective cosmic density field reconstruction with convolutional neural network

Using Bremsstrahlung for Electron Density Estimation and Correction in East Tokamak

Fast equilibrium reconstruction by deep learning on EAST tokamak

Neutron penumbral image reconstruction with a convolution neural network using fast Fourier transform

Electrical Impedance Tomography Image Reconstruction With Attention-Based Deep Convolutional Neural Network

One-dimensional Convolutional Neural Network (1D-CNN) Image Reconstruction for Electrical Impedance Tomography.

Image Super-resolution Inspired Electron Density Prediction

Convolutional network learning of self-consistent electron density via grid-projected atomic fingerprints

A Spectral Clustering Based Method for Automatic Plasma Pattern Identification in Electron Cyclotron Emission Imaging on EAST Tokamak

Shape Reconstruction With Multiphase Conductivity for Electrical Impedance Tomography Using Improved Convolutional Neural Network Method