Have a personal or library account? Click to login
RESE-CNN: Residual Squeeze-and-Excitation Network for High-Contrast Optical Tomography Reconstruction Cover

RESE-CNN: Residual Squeeze-and-Excitation Network for High-Contrast Optical Tomography Reconstruction

Measurement Science Review
Volume 25 (2025): Issue 2 (April 2025)
By: Yingkuang Zhu,  Zhenhua Pan,  Huajun Li,  Jianyang Chen and  Yihao Sheng  
Open Access
|Jun 2025

Figures & Tables

Fig. 1.

OT system.
OT system.

Fig. 2.

Tomography system and optical sensor. (a) Tomography system. (b) Optical sensor.
Tomography system and optical sensor. (a) Tomography system. (b) Optical sensor.

Fig. 3.

Geometry of one laser beam through the tomography field.
Geometry of one laser beam through the tomography field.

Fig. 4.

Mesh grid, beam arrangement, and sensitive map. (a) Mesh grid. (b) Beam arrangement. (c) Sensitive field.
Mesh grid, beam arrangement, and sensitive map. (a) Mesh grid. (b) Beam arrangement. (c) Sensitive field.

Fig. 5.

Main structure of RESE-CNN.
Main structure of RESE-CNN.

Fig. 6.

Substructure of the blocks. (a) Block in down sampling module. (b) Block in up sampling module.
Substructure of the blocks. (a) Block in down sampling module. (b) Block in up sampling module.

Fig. 7.

Structure of the SE attention.
Structure of the SE attention.

Fig. 8.

Simulated absorption coefficient distributions and reconstruction results in a 128 × 128 pixel grid.
Simulated absorption coefficient distributions and reconstruction results in a 128 × 128 pixel grid.

Fig. 9.

Results of one-bubble distribution in a 128 × 128 pixel grid.
Results of one-bubble distribution in a 128 × 128 pixel grid.

Fig. 10.

Results of two-bubble distribution in a 128 × 128 pixel grid.
Results of two-bubble distribution in a 128 × 128 pixel grid.

Fig. 11.

Results of three-bubble distribution in a 128 × 128 pixel grid.
Results of three-bubble distribution in a 128 × 128 pixel grid.

Fig. 12.

Results of four-bubble distribution in a 128 × 128 pixel grid.
Results of four-bubble distribution in a 128 × 128 pixel grid.

Fig. 13.

Comparison between the results of traditional algorithms and sensitive field.
Comparison between the results of traditional algorithms and sensitive field.

Fig. 14.

RE and CC of different distributions.
RE and CC of different distributions.

Fig. 15.

Results of the experiments.
Results of the experiments.

Reconstruction time of different methods_

MethodAverage reconstruction time (s)
LBP3.796 × 10−5
Landweber2.846
U-Net7.627 × 10−3
RESE-CNN7.687 × 10−3

Average RE and CC of different methods_

MethodAverage REAverage CC
LBP0.91520.4656
Landweber0.67270.7256
U-Net0.35750.9341
RE-CNN0.21640.9772
RESE-CNN0.18730.9776
DOI: https://doi.org/10.2478/msr-2025-0010 | Journal eISSN: 1335-8871
Journal RSS Feed
Language: English
Page range: 72 - 82
Submitted on: Jun 20, 2024
|
Accepted on: Apr 23, 2025
|
Published on: Jun 7, 2025
Published by: Slovak Academy of Sciences, Institute of Measurement Science
In partnership with: Paradigm Publishing Services
Publication frequency: Volume open
Keywords:
optical tomography,
deep learning,
reconstruction algorithm
Related subjects:
Engineering,
Electrical engineering,
Control engineering, metrology and testing

© 2025 Yingkuang Zhu, Zhenhua Pan, Huajun Li, Jianyang Chen, Yihao Sheng, published by Slovak Academy of Sciences, Institute of Measurement Science
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

Previous articleVolume 25 (2025): Issue 2 (April 2025)Next article