International Society for Industrial Process Tomography

An Inclusion Boundary Reconstruction Method Using Electrical Impedance and Ultrasound Reflection Dual-Modality Tomography

Guanghui Liang, Shangjie Ren^*, Feng Dong

Tianjin Key Laboratory of Process Measurement and Control

School of Electrical and Information Engineering, Tianjin University, Tianjin, P.R. China

^*Email: rensjie@tju.edu.cn

ABSTRACT

Inclusion detection problems are normally seen in industrial processes and accurately solving the problem is important for the optimization and control of the industrial production. Electrical Impedance Tomography (EIT) is a no-invasive technique to image the conductivity distribution within a closed domain, and it’s promising to solving the inclusion detection problems. However, due to the inherent ill-posedness and nonlinearity, EIT usually suffers the problem of low spatial resolution. To improve the problem, an ultrasound constrained EIT is proposed to directly reconstruct the geometric configuration of the inclusions inside the observation domain. In the proposed method, the shape of the target inclusion is parameterized as a set of shape coefficients by the radial shape model. Then, the dual-modality inclusion shape reconstruction model is built in the Bayesian statistics framework. In the framework, the likelihood model is derived from EIT forward problem, and the prior model is built from Ultrasound Reflection Tomography (URT). The inverse problem is to estimate the posterior distribution of the unknown shape coefficients conditioned on the EIT and URT measurements. Final, the Maximum a Posterior (MAP) estimate with the conventional error model is used to solve the dual-modality shape inversion problem and obtain the optimal shape coefficients. The numerical tests are carried out to evaluate the proposed method. Comparing with the traditional single modality EIT method, the proposed dual-modality shape reconstruction method has a higher accuracy and is more robust to the EIT measurement noise.

Keywords Dual-Modality Tomography, Electrical Impedance Tomography, Inclusion Boundary Reconstruction, Statistical Inverse, Ultrasound Reflection Tomography.

Industrial Application: Image Reconstruction.