Particle Swarm Optimization Technique For Elliptic Region Boundary Estimation In Electrical Impedance Tomography

A non-iterative linear inverse solution for the block approach in EIT

“…For this study RMSE of conductivity has been reported between 10 −2 to 10 −1 [23]. Using shape estimation of regions of known resistivities based on extended Kalman filtering for dynamic EIT results in RMSE from 10 −2 to 10 −1 [24]. Table 2 represents these results.…”

“…However in some papers RMSE has been adopted as a comparison factor [19,[22][23][24][25][26]. For example 10 −1 to 10 Shape estimation by EKF for dynamic EIT [24] 10 −2 to 10 −1 Employing IMM, EKF and CCEKF for dynamic EIT [23] 10 −2 to 10 −1 Neural networks and front point approach [25] 10 −2 to 10 −1 Interpolation of front points method [22] 10 −5 to 10 −3 Non-linear block method [19] 10 −5 to 10 −4 Proposed linear block method (this paper) 10 −8 to 10 −7 in a work conducted by Kim for 2D EIT inverse problem, applying interpolation of front points method for approximation of regions of inner object results in RMS on order of 10 −5 to 10 −3 for reconstructed image [22]. Also, using neural networks and front point approach for estimation of 2D EIT image results 10 −2 to 10 −1 in RMSE [25].…”

Boundary Estimation Techniques in Two-phase Flows using Electrical Impedance Tomography: A Review

“…Jeon et al [22] have used Multi-layered neural network (MNN) to estimate the Fourier coefficients as it is conceptually simple, easier in implementation and the principal advantage is that it does not require computation of the Jacobian matrix. Recently, derivative free methods such as pattern search method [14] and evolutionary algorithms such as particle swarm optimization [23] are applied for the closed boundary estimation of void regions in flow field. These methods offer better performance when compared to mNR and are robust to initial condition and noise.…”