Fuzzy clustering based ET image fusion

Yue, Shihong; Wu, Teresa; Pan, Jiluan; Wang, Huaxiang

doi:10.1016/j.inffus.2012.09.004

Cited by 32 publications

(18 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Even with the LBP algorithm, it is capable to provide satisfactory results. Recently, more sophisticated algorithms like the nonlinear inversion technique and the fuzzy clustering method have been developed, which were proved to be efficient in ECT reconstruction to reveal complicated permittivity patterns [17], [18]. Future work will focus on using such advanced image reconstruction algorithms to achieve better results.…”

Section: Discussionmentioning

confidence: 99%

Liquid Film Thickness Estimation using Electrical Capacitance Tomography

Cui¹,

Yang²,

Sun³

et al. 2014

Measurement Science Review

Self Cite

View full text Add to dashboard Cite

In air/oil lubrication systems, the flow parameters, e.g., flow pattern, liquid film thickness, and air/oil flow rate, are of great importance to the transportation efficiency. In most cases, the on-going two-phase flow is annular flow with the oil moving along the tube wall and the air travelling at high speed in the center. This usually results in the formation of a thin oil film, the thickness of which is a key parameter determining the efficiency of the lubrication system. As the oil film thickness of the on-going air/oil flow varies dynamically, there is actually no applicable method for a non-intrusive test. In this paper, the use of electrical capacitance tomography (ECT) to investigate the air/oil flow has been studied. Capacitance measurements are made from an externally mounted electrode array in a non-invasive and non-intrusive manner. Both average and distributed oil film thicknesses can be calculated from the reconstructed ECT images. Simulation and experimental results show that the ECT technique can provide satisfactory results of online oil film thickness estimation.

show abstract

Section: Discussionmentioning

confidence: 99%

Liquid Film Thickness Estimation using Electrical Capacitance Tomography

Cui¹,

Yang²,

Sun³

et al. 2014

Measurement Science Review

Self Cite

View full text Add to dashboard Cite

show abstract

“…Recently, one efficient and original EIT imaging algorithm [16], termed as FC-EIT, has been proposed. FC-EIT first maps K measuring data of pixel j after K excitations into a K -dimensional vector as

\begin{matrix} {(v_{j, 1} v_{j, 2}, \dots, v_{j, K})}^{T}, \forall j = 1, 2, \dots, n . \end{matrix}

All vectors associated with all pixels consist of a set X and then all vectors in X are partitioned into c clusters by the fuzzy c -means (FCM) algorithm [17].…”

Section: Eit Imaging Principlementioning

confidence: 99%

An Effective Measured Data Preprocessing Method in Electrical Impedance Tomography

Yue

Wang

et al. 2014

The Scientific World Journal

Self Cite

View full text Add to dashboard Cite

As an advanced process detection technology, electrical impedance tomography (EIT) has widely been paid attention to and studied in the industrial fields. But the EIT techniques are greatly limited to the low spatial resolutions. This problem may result from the incorrect preprocessing of measuring data and lack of general criterion to evaluate different preprocessing processes. In this paper, an EIT data preprocessing method is proposed by all rooting measured data and evaluated by two constructed indexes based on all rooted EIT measured data. By finding the optimums of the two indexes, the proposed method can be applied to improve the EIT imaging spatial resolutions. In terms of a theoretical model, the optimal rooting times of the two indexes range in [0.23, 0.33] and in [0.22, 0.35], respectively. Moreover, these factors that affect the correctness of the proposed method are generally analyzed. The measuring data preprocessing is necessary and helpful for any imaging process. Thus, the proposed method can be generally and widely used in any imaging process. Experimental results validate the two proposed indexes.

show abstract

“…There are at least three advantages of ERT over the existing tomographic techniques, such as low-cost, rapid response, noninvasion and so on. In process tomography area, ERT has succeeded in several typical areas, such as flow pattern recognition, many found many applications, gas/liquid 978-1-4799-0260-6/13/$31.00 ©2013 IEEE measurement in pipelines [3][4][5] and the estimation of granular shapes in fluidized beds [6,7].…”

Section: Introductionmentioning

confidence: 99%

“…In terms of the vector and matrix form, the equation of Eq. (2) can be rewritten as (3) where U and S are the voltage vector and sensitivity coefficient matrix, respectively, and a is the conductivity vector that consist of the grey levels of all pixels. Consequently, the task of ERT imaging is to solve the unknown a from the known U, as explained below (4) But there doesn't exist a direct analytical solution of Eq.…”

mentioning

confidence: 99%

An ERT imaging algorithm based on Nelder-mead simplex method

Yue

Zhou

2013

2013 International Conference on Machine Learning and Cybernetics

Self Cite

View full text Add to dashboard Cite

Electrical resistance tomography (ERT) techniques have increasing gained importance during the past several decades as a process visualization tool. Most of the existing ERT methods work in term of an optimization process and assume the optimized objective function be differentiable. But this assumption doesn't hold in noisy conditions and leads to an infeasible optimization process. In this paper, a new ERT imaging method based on NeIder-mead simplex algorithm is proposed. The key feature of the algorithm can efficiently work without the differentiability constraint. This strategy can efficiently enhance spatial resolution with less calculation than the existing algorithms. A group of experimentsshow that the proposed method can work better than two existing ET imaging methods in complex conditions with noisy data.

show abstract

Fuzzy clustering based ET image fusion

Cited by 32 publications

References 28 publications

Liquid Film Thickness Estimation using Electrical Capacitance Tomography

Liquid Film Thickness Estimation using Electrical Capacitance Tomography

An Effective Measured Data Preprocessing Method in Electrical Impedance Tomography

An ERT imaging algorithm based on Nelder-mead simplex method

Contact Info

Product

Resources

About