ECT image reconstruction based on one-dimensional convolution neural network optimized with particle swarm optimization algorithm

Min Ma; Tian Zeng

doi:10.1117/12.2652206

10 November 2022 ECT image reconstruction based on one-dimensional convolution neural network optimized with particle swarm optimization algorithm

Min Ma, Tian Zeng

Author Affiliations +

Proceedings Volume 12331, International Conference on Mechanisms and Robotics (ICMAR 2022); 123314O (2022) https://doi.org/10.1117/12.2652206
Event: International Conference on Mechanisms and Robotics (ICMAR 2022), 2022, Zhuhai, China

Abstract

In order to solve the problem of unqualitative, ill-conditioned and nonlinear in Electrical Capacitance Tomography (ECT), an image reconstruction based on one-dimensional convolution neural network optimized with particle swarm optimization algorithm (PSO-1DCNN) was proposed. A one-dimensional convolutional neural network (1DCNN) was built to learn the mapping relationship between capacitance and permittivity according to the data characteristics of ECT image reconstruction, and a particle swarm optimization algorithm (PSO) was used to search the global optimal hyperparameters of 1DCNN. The experimental results show that compared with Landweber and Tikhonov regularization algorithms,1DCNN algorithm can accurately predict the location and size of the bubble type with fewer artifacts. The correlation coefficient is up to 0.9253, and the image error is as low as 0.0863. The correlation coefficient of PSO-1DCNN algorithm is up to 0.9626. The image error is as low as 0.0357, and the image reconstruction effect is better.

Citation Download Citation

Min Ma and Tian Zeng "ECT image reconstruction based on one-dimensional convolution neural network optimized with particle swarm optimization algorithm", Proc. SPIE 12331, International Conference on Mechanisms and Robotics (ICMAR 2022), 123314O (10 November 2022); https://doi.org/10.1117/12.2652206

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available