基于卷积神经网络及有限元仿真的电容层析成像图像重建

doi:10.16182/j.issn1004731x.joss.20-0849

摘要/Abstract

摘要：

为求解电容层析成像（electrical capacitance tomography，ECT）图像重建非线性病态逆问题，提出基于一维卷积神经网络（one-dimensional convolutional neural network，1D CNN）的电容层析成像图像重建算法。通过1D CNN建立了ECT系统独立测量值与重建图像灰度值之间的非线性映射关系。采用有限元仿真软件获得6种典型流型的随机分布样本，完成了1D CNN网络的训练及测试，进行了静态实验，并与常用的线性反投影、Landweber迭代算法进行比较。仿真及静态实验结果均表明：基于1D CNN算法的ECT重建图像质量得到明显提高，具有较好的泛化能力及实时性。

关键词: 两相流, 电容层析成像, 一维卷积神经网络, 图像重建, 有限元仿真

Abstract:

In order to resolve the nonlinear and ill-posed inverse problem of the image reconstruction of electrical capacitance tomography (ECT), an image reconstruction algorithm based on one-dimensional convolutional neural network (1D CNN) is presented. The nonlinear mapping relationship between the independent measurement value of ECT system and the gray value of reconstructed image is established by 1D CNN. Six typical flow regimes with random distribution are obtained by the finite element simulation software and a 1D CNN is successfully trained. Simulation and static experiments are carried out and the reconstructed images using linear back projection, Landweber iterative algorithm and 1D CNN are compared. Experimental results show that the 1D CNN algorithm has good generalization ability and the quality of reconstructed images by 1D CNN is obviously improved compared with the other two algorithms.

Key words: two-phase flow, electrical capacitance tomography, one-dimensional convolutional neural network, image reconstruction, finite element simulation

中图分类号:

TP391

张立峰, 王会忍. 基于卷积神经网络及有限元仿真的电容层析成像图像重建[J]. 系统仿真学报, 2022, 34(4): 712-718.

Lifeng Zhang, Huiren Wang. Image Reconstruction of Electrical Capacitance Tomography Based on Convolutional Neural Network and Finite Element Simulation[J]. Journal of System Simulation, 2022, 34(4): 712-718.

图/表 11

图1

图2

图3

图4

图5

图6

表1

表2

图7

表3

图8

参考文献 16

1	Lei J, Liu S. Dynamic Inversion Approach for Electrical Capacitance Tomography [J]. IEEE Transaction on Instrumentation and Measurement(S0018-9456), 2013, 62(11): 3035-3049.
2	王丕涛, 王化祥, 孙犇渊. 基于内阵列电极的ECT/ERT双模态成像[J]. 天津大学学报(自然科学与工程技术版), 2015, 48(4): 311-316.
	Wang Pitao, Wang Huaxiang, Sun Benyuan. ECT/ERT Dual-Modality Tomography Based on Inner Electrode Array[J]. Journal of Tianjin University(Science and Technology), 2015, 48(4): 311-316.
3	王小鑫, 王博, 陈阳正, 等. 基于电容层析成像技术重构图像的两相流流型识别[J]. 计量学报, 2020, 41(8): 942-946.
	Wang Xiaoxin, Wang Bo, Chen Yangzheng, et al. Two-phase Flow Pattern Recognition Based on Electrical Capacitance Tomography Reconstructed Images [J]. Acta Metrologica Sinica, 2020, 41(8): 942-946.
4	张立峰, 朱炎峰. 极限学习机在电容层析成像中的应用[J]. 电测与仪表, 2020, 57(9): 146-150.
	Zhang Lifeng, Zhu Yanfeng. Application of Extreme Learning Machine in Electrical Capacitance Tomography [J]. Electrical Measurement ＆ Instrumentation, 2020, 57(9): 146-150.
5	吴新杰, 李红玉, 梁南南. 卷积神经网络在 ECT 图像重建上的应用[J]. 辽宁大学学报, 2018, 45(1): 29-33.
	Wu Xinjie, Li Hongyu, Liang Nannan. Application of Convolutional Neural Network in ECT Image Reconstruction [J]. Journal of Liaoning University, 2018, 45(1): 29-33.
6	张耘. 电容层析成像系统成像算法研究[D]. 哈尔滨: 哈尔滨理工大学, 2015.
	Zhang Yun. Research on Imaging Algorithm of Electrical Capacitance Tomography System[D]. Harbin: Harbin University of Technology, 2015.
7	孙先亮, 李健, 韩哲哲, 等. 基于数据驱动的卷积神经网络电容层析成像图像重建[J]. 化工学报, 2020, 71(5): 2004-2016.
	Sun Xianliang, Li Jian, Han Zhezhe, et al. Image Reconstruction of Electrical Capacitance Tomography Based on Data Driven Convolution Neural Network [J]. Journal of Chemical Engineering, 2020, 71(5): 2004-2016.
8	Meribout M, Saied I M. Real-Time Two-Dimensional Imaging of Solid Contaminants in Gas Pipelines Using an Electrical Capacitance Tomography System[J]. IEEE Transactions on Industrial Electronics(S0278-0046), 2017, 64(5): 3989-3996.
9	陈宇, 夏宗基, 周雨佳. 基于修正稀疏拟牛顿的电容层析成像重建算法[J]. 系统仿真学报, 2019, 31(5): 820-821.
	Chen Yu, Xia Zongji, Zhou Yujia. Modified Sparse Quasi-newton Algorithm for Electrical Capacitance Tomography System[J]. Journal of System Simulation, 2019, 31(5): 820-821.
10	张立峰, 李佳, 田沛. Kalman滤波在电容层析成像图像重建中的应用[J]. 计量学报, 2017, 38(3): 316-318.
	Zhang Lifeng, Li Jia, Tian Pei. Application of Kalman Filter for Image Reconstruction of Electrical Capacitance Tomography [J]. Acta Metrologica Sinica, 2017, 38(3): 316-318.
11	刘诗源. 电容层析成像系统关键技术研究与设计[D]. 南昌: 华东交通大学, 2017.
	Liu Shiyuan. Research and Design of Key Technology for Capacitance Tomography System[D]. Nanchang: East China Jiaotong University, 2017.
12	马敏, 郭琪, 闫超齐. 基于广义正则化的ECT图像重建[J]. 系统仿真学报, 2017, 29(8): 1851-1853.
	Ma Min, Guo Qi, Yan Chaoqi. ECT Image Reconstruction Algorithm Based on Generalized Regularization[J]. Journal of System Simulation, 2017, 29(8): 1851-1853.
13	Li F, Tan C, Dong F, et al. V-Net Deep Imaging Method for Electrical Resistance Tomography [J]. IEEE Sensors Journal(S1530-437X), 2020, 20(12): 6460-6468.
14	于波, 方业全, 刘闽, 等. 基于深度卷积神经网络的图像重建算法[J]. 计算机系统应用, 2018, 27(9): 170-175.
	Yu Bo, Fang Yequan, Liu Min, et al. Image Reconstruction Algorithm Based on Deep Convolution Neural Network [J]. Computer Systems & Applications, 2018, 27(9): 170-175.
15	颜华, 许鑫, 孙延辉. 带阈值滤波的迭代法三维ECT图像重建[J]. 沈阳工业大学学报, 2014, 36(6): 671-675.
	Yan Hua, Xu Xin, Sun Yanhui. 3D ECT Image Reconstruction by Iterative Method with Threshold Filtering[J]. Journal of Shenyang University of Technology, 2014, 36(6): 671-675.
16	Tan C, Dong F, Masahiro T. Image Reconstruction Based on Convolutional Neural Network for Electrical Resistance Tomography [J]. IEEE Sensors Journal(S1530-437X), 2019, 19(1): 196-203.

流型	指标	LBP	Landweber	1D CNN
1	RE	0.925 6	0.553 3	0.179 8
1	CC	0.587 7	0.830 0	0.983 1
2	RE	0.913 0	0.677 7	0.263 2
2	CC	0.422 0	0.753 6	0.963 4
3	RE	0.909 8	0.672 2	0.344 8
3	CC	0.429 1	0.758 1	0.930 0
4	RE	0.739 7	0.545 6	0.000 1
4	CC	0.571 1	0.771 2	0.999 9
5	RE	0.545 2	0.231 6	0.064 6
5	CC	0.859 8	0.900 3	0.992 5
6	RE	1.063 9	0.549 3	0.125 0
6	CC	0.679 4	0.717 8	0.986 8

算法	重建时间
LBP	0.005
Landweber	0.686
1D CNN	0.438

流型	指标	LBP	Landweber	1D CNN
1	RE	0.887 9	0.587 6	0.539 6
1	CC	0.517 9	0.796 0	0.840 5
2	RE	0.863 7	0.472 5	0.402 4
2	CC	0.607 4	0.868 5	0.904 0
3	RE	0.939 1	0.555 3	0.446 3
3	CC	0.433 9	0.823 6	0.877 2
4	RE	0.848 3	0.718 0	0.652 3
4	CC	0.433 4	0.655 6	0.703 6

[1]	仝卫国, 曾世超, 张立峰, 侯哲, 郭佳跃. 基于深度残差神经网络的电阻层析成像及流型辨识方法[J]. 系统仿真学报, 2022, 34(9): 2028-2036.
[2]	仝卫国, 庞雪纯, 朱赓宏. 基于卷积神经网络的气液两相流流型识别方法[J]. 系统仿真学报, 2021, 33(4): 883-891.
[3]	汪送, 战仁军, 段雄义. 橡皮弹胸部钝击损伤的有限元法评估[J]. 系统仿真学报, 2020, 32(9): 1762-1770.
[4]	陈宇, 夏宗基, 周雨佳. 基于修正稀疏拟牛顿的电容层析成像重建算法[J]. 系统仿真学报, 2019, 31(5): 819-827.
[5]	魏文礼, 娄威立, 李盼盼, 刘玉玲. 倒伞形表面曝气搅拌槽内气液两相流的模拟研究[J]. 系统仿真学报, 2019, 31(4): 696-701.
[6]	何永勃, 董玉珊, 薛荣荣. 基于交互式递归分析的两相流流型识别方法[J]. 系统仿真学报, 2019, 31(4): 720-726.
[7]	童成彪, 周志雄, 阎秋霞. 固定锥形阀的补气及排放系数研究[J]. 系统仿真学报, 2018, 30(8): 3188-3197.
[8]	王艳, 朱新庆, 胡捷飞, 崔西民. 四辊卷板机连续滚弯成形工艺的数值模拟研究[J]. 系统仿真学报, 2018, 30(5): 1772-1780.
[9]	王进峰, 赵久兰, 储开宇. SiCp/Al复合材料切削力的仿真研究[J]. 系统仿真学报, 2018, 30(4): 1566-1571.
[10]	鲁正平, 郭旭东, 许堂成, 秦锐华. 电磁-永磁式遥控施药胶囊驱动装置优化设计[J]. 系统仿真学报, 2018, 30(2): 747-752.
[11]	马敏, 郭琪, 闫超奇. 基于广义正则化的ECT图像重建算法[J]. 系统仿真学报, 2017, 29(8): 1851-1858.
[12]	朱胜伟, 张丽慧, 洪滔, 章城. 液动压悬浮抛光固液两相流数值模拟[J]. 系统仿真学报, 2017, 29(12): 2993-3000.
[13]	律方成, 郭云翔, 李鹏. 基于轻量化需求的大功率中频变压器优化设计[J]. 系统仿真学报, 2017, 29(1): 154-161.
[14]	莫丽, 杨琪, 王玉梅. 水力加载器内部汽蚀数值模拟及失效分析[J]. 系统仿真学报, 2016, 28(6): 1432-1438.
[15]	岳彩旭, 蔡春彬, 黄翠, 刘二亮. 切削加工过程有限元仿真研究的最新进展[J]. 系统仿真学报, 2016, 28(4): 815-825.