基于非凸与不可分离正则化算法的电容层析成像图像重建

doi:10.11949/0438-1157.20240001

摘要/Abstract

摘要：

搅拌器内两相混合是化工生产中常见的现象，电容层析成像(ECT)技术主要对两相分布进行可视化重构，以达到监测的目的。受稀疏贝叶斯学习的启发，提出了一种非凸与不可分离正则化(NNR)算法重建ECT图像。在稀疏先验的基础上引入矩阵低秩特性，采用最大后验估计在潜在空间中提出一个新的优化问题，利用对偶变量将潜在空间的目标函数映射到原始空间进行迭代求解，用来恢复同时稀疏与低秩的矩阵。与凸近似L¹范数相比，NNR算法可获得更准确的重建图像，同时比非凸可分离方法更容易收敛到全局最优解。为验证NNR算法的重建效果，通过数值仿真与静态实验的方法分别与其他5种算法进行重建对比。结果表明：NNR算法可以有效减少重建伪影，提升中心物体的重建质量，为搅拌器内两相分布提供了高质量的重建算法。

关键词: 电容层析成像, 图像重建, 非凸不可分离正则化, 稀疏-低秩模型, 两相混合

Abstract:

Two-phase mixing in a stirrer is a common phenomenon in chemical production. Electrical capacitance tomography (ECT) technology mainly visually reconstructs the distribution of the two phases for monitoring purposes. Inspired by sparse Bayesian learning, a non-convex and nonseparable regularization (NNR) algorithm is proposed to reconstruct ECT images. The low-rank characteristics of the matrix are introduced on the basis of the sparse prior, and a new optimization problem is proposed in the latent space by using the maximum posterior estimation. Dual variables are used to map the objective function of the latent space to the original space for an iterative solution, which is used to restore the simultaneous sparse and low-rank matrices. Compared with the convex approximation L¹ norm, the NNR algorithm can obtain more accurate reconstruction images, and it is easier to converge to the global optimal solution than the non-convex separable method. To verify the reconstruction effect of the NNR algorithm, the reconstruction was compared with the other five algorithms through numerical simulation and static experiments. The results show that the NNR algorithm can effectively reduce reconstruction artifacts, improve the reconstruction quality of the central object, and provide a high-quality reconstruction algorithm for the two-phase distribution in the stirrer.

Key words: electrical capacitance tomography, image reconstruction, non-convex and nonseparable regularization, sparse-low-rank model, two-phase mixture

中图分类号:

TB 937

李宁, 朱朋飞, 张立峰, 卢栋臣. 基于非凸与不可分离正则化算法的电容层析成像图像重建[J]. 化工学报, 2024, 75(3): 836-846.

Ning LI, Pengfei ZHU, Lifeng ZHANG, Dongchen LU. Image reconstruction of electrical capacitance tomography based on non-convex and nonseparable regularization algorithm[J]. CIESC Journal, 2024, 75(3): 836-846.

图/表 10

参考文献 33

1	Zhang L F, Zhang S J. Analysis and identification of gas-liquid two-phase flow pattern based on multi-scale power spectral entropy and pseudo-image encoding[J]. Energy, 2023, 282: 128835.
2	刘马林. 流化床-化学气相沉积技术在先进核燃料制备中的应用进展[J]. 化工进展, 2019, 38(4): 1646-1653.
	Liu M L. Research activities on FB-CVD technology application in advanced nuclear fuel fabrication[J]. Chemical Industry and Engineering Progress, 2019, 38(4): 1646-1653.
3	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 X L, Li J, Han Z Z, et al. Data-driven image reconstruction of electrical capacitance tomography based on convolutional neural network[J]. CIESC Journal, 2020, 71(5): 2004-2016.
4	Warsito W, Fan L S. ECT imaging of three-phase fluidized bed based on three-phase capacitance model[J]. Chemical Engineering Science, 2003, 58(3/4/5/6): 823-832.
5	郑榕根, 周骛, 孔庆恩, 等. 燃烧煤粉颗粒的粒度速度测量方法[J]. 煤炭学报, 2017, 42(6): 1579-1584.
	Zheng R G, Zhou W, Kong Q E, et al. Particle size velocity measurement method for burning pulverized coal particles[J]. Journal of China Coal Society, 2017, 42(6): 1579-1584.
6	罗琴, 赵银峰, 叶茂, 等. 电容层析成像在气固流化床测量中的应用[J]. 化工学报, 2014, 65(7): 2504-2512.
	Luo Q, Zhao Y F, Ye M, et al. Application of electrical capacitance tomography for gas-solid fluidized bed measurement[J]. CIESC Journal, 2014, 65(7): 2504-2512.
7	Li Y, Yang W Q, Xie C G, et al. Gas/oil/water flow measurement by electrical capacitance tomography[J]. Measurement Science and Technology, 2013, 24(7): UNSP 074001.
8	Cui Z Q, Wang Q, Xue Q, et al. A review on image reconstruction algorithms for electrical capacitance/resistance tomography[J]. Sensor Review, 2016, 36(4): 429-445.
9	Yang W Q, Peng L H. Image reconstruction algorithms for electrical capacitance tomography[J]. Measurement Science and Technology, 2003, 14(1): 1-13.
10	Li Y, Yang W Q. Image reconstruction by nonlinear Landweber iteration for complicated distributions[J]. Measurement Science and Technology, 2008, 19(9): 094014.
11	Chu H V. Performance of the thresholding greedy algorithm with larger greedy sums[J]. Journal of Mathematical Analysis and Applications, 2023, 525(1): 127126.
12	马敏, 郭鑫, 于洁. 改进正则化半阈值算法的ECT图像重建[J]. 仪器仪表学报, 2022, 43(5): 110-119.
	Ma M, Guo X, Yu J. ECT image reconstruction based on improved regularized half threshold algorithm[J]. Chinese Journal of Scientific Instrument, 2022, 43(5): 110-119.
13	江鹏, 彭黎辉, 陆耿, 等. 基于贝叶斯理论的电容层析成像图像重建迭代算法[J]. 中国电机工程学报, 2008, 28(11): 65-71.
	Jiang P, Peng L H, Lu G, et al. Iterative image reconstruction algorithm based on Bayesian theorem for electrical capacitance tomography[J]. Proceedings of the CSEE, 2008, 28(11): 65-71.
14	de Moura H L, Pipa D R, do Nascimento Wrasse A, et al. Image reconstruction for electrical capacitance tomography through redundant sensitivity matrix[J]. IEEE Sensors Journal, 2017, 17(24): 8157-8165.
15	Zhang L F, Dai L. Image reconstruction of electrical capacitance tomography based on an efficient sparse Bayesian learning algorithm[J]. IEEE Transactions on Instrumentation and Measurement, 2022, 71: 1-14.
16	Kabashima Y, Wadayama T, Tanaka T. A typical reconstruction limit for compressed sensing based on L_p -norm minimization[J]. Journal of Statistical Mechanics: Theory and Experiment, 2009(9): 1-14.
17	Selesnick I W, Bayram İ. Enhanced sparsity by non-separable regularization[J]. IEEE Transactions on Signal Processing, 2016, 64(9): 2298-2313.
18	Wipf D P, Rao B D, Nagarajan S. Latent variable Bayesian models for promoting sparsity[J]. IEEE Transactions on Information Theory, 2011, 57(9): 6236-6255.
19	Chen W. Simultaneously sparse and low-rank matrix reconstruction via nonconvex and nonseparable regularization[J]. IEEE Transactions on Signal Processing, 2018, 66(20): 5313-5323.
20	张立峰, 宋亚杰. 基于梯度投影稀疏重建算法的电容层析成像图像重建[J]. 计量学报, 2019, 40(4): 631-635.
	Zhang L F, Song Y J. Image reconstruction for electrical capacitance tomography based on Barzilai-Borwein gradient projection for sparse reconstruction algorithm[J]. Acta Metrologica Sinica, 2019, 40(4): 631-635.
21	Ye J M, Wang H G, Yang W Q. Image recovery for electrical capacitance tomography based on low-rank decomposition[J]. IEEE Transactions on Instrumentation and Measurement, 2017, 66(7): 1751-1759.
22	Donoho D L. Compressed sensing[J]. IEEE Transactions on Information Theory, 2006, 52(4): 1289-1306.
23	Tipping M E. Sparse Bayesian learning and the relevance vector machine[J]. Journal of Machine Learning Research, 2001, 1: 211-244.
24	Wang L, Zhao L F, Rahardja S, et al. Alternative to extended block sparse Bayesian learning and its relation to pattern-coupled sparse Bayesian learning[J]. IEEE Transactions on Signal Processing, 2018, 66(10): 2759-2771.
25	Xin B, Wang Y Z, Gao W, et al. Exploring algorithmic limits of matrix rank minimization under affine constraints[J]. IEEE Transactions on Signal Processing, 2016, 64(19): 4960-4974.
26	董峰, 赵佳, 许燕斌, 等. 用于电阻层析成像的快速自适应硬阈值迭代算法[J]. 天津大学学报(自然科学与工程技术版), 2015, 48(4): 305-310.
	Dong F, Zhao J, Xu Y B, et al. A fast adaptive iterative hard threshold algorithm for electrical resistance tomography[J]. Journal of Tianjin University(Science and Technology), 2015, 48(4): 305-310.
27	Saperas-Riera J, Mateu-Figueras G, Martín-Fernández J A. Lasso regression method for a compositional covariate regularised by the norm L ¹ pairwise logratio[J]. Journal of Geochemical Exploration, 2023, 255: 107327.
28	Yang Y P, Yang Z Z, Li J L, et al. Foreground-background separation via generalized nuclear norm and structured sparse norm based low-rank and sparse decomposition[J]. IEEE Access, 2020, 8: 84217-84229.
29	杨钢, 王玉涛, 邵富群, 等. 电容层析成像图像重建中的迭代算法[J]. 仪器仪表学报, 2006, 27(12): 1591-1594.
	Yang G, Wang Y T, Shao F Q, et al. Iterative methods in ECT image reconstruction[J]. Chinese Journal of Scientific Instrument, 2006, 27(12): 1591-1594.
30	Ji S H, Xue Y, Carin L. Bayesian compressive sensing[J]. IEEE Transactions on Signal Processing, 2008, 56(6): 2346-2356.
31	张立峰, 王化祥. 一种新的电容层析成像电极组合激励测量模式[J]. 化工学报, 2012, 63(3): 860-865.
	Zhang L F, Wang H X. A new combined-electrode exciting-measuring mode for electrical capacitance tomography[J]. CIESC Journal, 2012, 63(3): 860-865.
32	Zhang L F, Liu J, Tian P. Six rotating electrodes exciting-measuring mode for electrical capacitance tomography system[C]// 2015 8th International Congress on Image and Signal Processing. IEEE, 2015: 1000-1004.
33	Ye J M, Wang H G, Yang W Q. Image reconstruction for electrical capacitance tomography based on sparse representation[J]. IEEE Transactions on Instrumentation and Measurement, 2015, 64(1): 89-102.

编辑推荐 0

Metrics

阅读次数

全文

100

HTML			PDF

最新录用	在线预览	正式出版	最新录用	在线预览	正式出版
0	0	17	6	0	77

来源	本网站	其他网站

次数	48	52
比例	48%	52%

摘要

139

最新录用	在线预览	正式出版

9	0	130

来源	本网站	其他网站

次数	31	108
比例	22%	78%

分布	IHT	L^0.5	L¹	Landweber	MSBL	NNR
a	0.4811	0.4599	0.4879	0.5412	0.9179	0.3766
b	0.9070	0.8207	0.8165	0.9825	0.9260	0.7309
c	0.7719	0.6908	0.6149	0.5595	0.9311	0.4285
d	0.6692	0.5451	0.6059	0.6792	0.9341	0.3842
e	0.8826	0.7275	0.7291	0.9043	0.8737	0.5670
f	0.8027	0.4123	0.4189	0.5876	0.9400	0.3557
g	0.7914	0.6982	0.6402	0.7667	0.9399	0.5067
h	0.7919	0.4133	0.4834	0.7566	0.9039	0.3946
i	0.8747	0.7166	0.7168	0.9593	0.8735	0.5682
j	0.7403	0.3552	0.5806	0.7853	0.9623	0.1175

分布	IHT	L^0.5	L¹	Landweber	MSBL	NNR
a	0.4811	0.4599	0.4879	0.5412	0.9179	0.3766
b	0.9070	0.8207	0.8165	0.9825	0.9260	0.7309
c	0.7719	0.6908	0.6149	0.5595	0.9311	0.4285
d	0.6692	0.5451	0.6059	0.6792	0.9341	0.3842
e	0.8826	0.7275	0.7291	0.9043	0.8737	0.5670
f	0.8027	0.4123	0.4189	0.5876	0.9400	0.3557
g	0.7914	0.6982	0.6402	0.7667	0.9399	0.5067
h	0.7919	0.4133	0.4834	0.7566	0.9039	0.3946
i	0.8747	0.7166	0.7168	0.9593	0.8735	0.5682
j	0.7403	0.3552	0.5806	0.7853	0.9623	0.1175

分布	IHT	L^0.5	L¹	Landweber	MSBL	NNR
a	0.8936	0.8915	0.8857	0.8734	0.3872	0.9299
b	0.4015	0.5610	0.5640	0.5757	0.3716	0.7525
c	0.6256	0.7851	0.8250	0.8258	0.3625	0.8969
d	0.7324	0.8503	0.8239	0.7248	0.3435	0.9150
e	0.4582	0.6802	0.6741	0.5526	0.4796	0.8278
f	0.5801	0.9060	0.9051	0.7898	0.3349	0.9424
g	0.5756	0.6909	0.7402	0.6070	0.3257	0.8501
h	0.5590	0.8966	0.8548	0.5921	0.4027	0.9066
i	0.4583	0.6815	0.6807	0.3916	0.4709	0.8227
j	0.6498	0.9152	0.7936	0.5882	0.2357	0.9899

分布	IHT	L^0.5	L¹	Landweber	MSBL	NNR
a	0.8936	0.8915	0.8857	0.8734	0.3872	0.9299
b	0.4015	0.5610	0.5640	0.5757	0.3716	0.7525
c	0.6256	0.7851	0.8250	0.8258	0.3625	0.8969
d	0.7324	0.8503	0.8239	0.7248	0.3435	0.9150
e	0.4582	0.6802	0.6741	0.5526	0.4796	0.8278
f	0.5801	0.9060	0.9051	0.7898	0.3349	0.9424
g	0.5756	0.6909	0.7402	0.6070	0.3257	0.8501
h	0.5590	0.8966	0.8548	0.5921	0.4027	0.9066
i	0.4583	0.6815	0.6807	0.3916	0.4709	0.8227
j	0.6498	0.9152	0.7936	0.5882	0.2357	0.9899

[1]	陈昭, 陈猛, 王江江, 常家兴, 刘马林. 非规则结构电容层析成像填补法测量的敏感场特性及重构算法改进[J]. 化工学报, 2020, 71(8): 3469-3479.
[2]	孙先亮, 李健, 韩哲哲, 许传龙. 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.
[3]	张立峰1，王化祥2. 一种新的电容层析成像电极组合激励测量模式[J]. 化工学报, 2012, 63(3): 860-865.
[4]	付飞飞, 许传龙, 王式民. 密相气力输送中气固两相流动特性多源信息分析[J]. 化工学报, 2012, 63(10): 3070-3079.
[5]	李虎, 杨道业, 程明霄. 螺旋表面极板电容式传感器的灵敏场分析 [J]. 化工学报, 2011, 62(8): 2292-2297.
[6]	孔祥明, 程振民. 涓流床中液体分布形态与滞后特性[J]. 化工学报, 2011, 62(6): 1515-1523.
[7]	王琦, 王化祥. 基于小波多分辨分析和最小交叉熵的快速CT图像重建算法 [J]. 化工学报, 2011, 62(1): 59-65.
[8]	程振民, 孔祥明, 曹鑫, 兰江安, 黄子宾. 涓流床反应器中流区过渡的气相渗透率表征 [J]. 化工学报, 2010, 61(7): 1770-1776.
[9]	马胜, 郭晓镭, 龚欣, 黄万杰, 陆海峰, 刘剀. 粉煤密相气力输送流型 [J]. 化工学报, 2010, 61(6): 1415-1422.
[10]	周宾;郝英立;王式民. 基于光学分层层析成像原理的火焰测温法 [J]. CIESC Journal, 2010, 61(3): 612-622.
[11]	杨道业;周宾;王式民. 电容层析成像在高压浓相煤粉气力输送中的应用 [J]. CIESC Journal, 2009, 60(4): 892-897.
[12]	吴晓. 柱塞气固两相流输送特性 [J]. CIESC Journal, 2009, 60(3): 608-614.
[13]	江鹏, 彭黎辉, 萧德云. 采用二阶导数阵作为正则化的电容成像图像重建算法 [J]. 化工学报, 2008, 59(2): 405-409.
[14]	李强伟;黄志尧;王保良;李海青. 基于ECT和蚂蚁算法的油气两相流空隙率在线测量 [J]. CIESC Journal, 2007, 58(1): 61-66.
[15]	沈熊; 彭涛. 激光散射图像可视化测量两相混合流场浓度分布 [J]. CIESC Journal, 2001, 52(7): 642-645.