ECVT Image Reconstruction Improvement with Sectorized Sensitivity Matrix

    P N Darma1, M R Baidillah1,2, A Saputra1, H Rosyad1, D Handoko1, Warsito P Taruno1
    7th International Symposium on Process Tomography, 1 – 3 September 2015,The Westin Bellevue, Dresden, Germany

    1 CTECH Labs Edwar Technology. Kav. Spectra 23BC, Alam Sutera, Kota Tangerang,
    Banten, 15144 Indonesia
    2 Graduate School of Engineering, Div. Artificial System Science. Chiba University,
    Chiba-shi, Inage-ku, Yayoi-cho, 1-33. 263-0022 Japan

    E-mail: panji.nursetia@c-techlabs.com

    Abstract:

    Electrical Capacitance Volume Tomography (ECVT) as one of tomography modalities based on electrical property measurement becomes more relevant for imaging of dielectric materials in industries as well as medicines. Sensitivity map is still used widely for solving inverse problem in electrical tomography today. The sensitivity map represents a region enclosed by the electrode pair which is sensitive to the change in the permittivity distribution. Due to the nonlinear characteristic of the electric field, an opposite pair of electrodes has region with lower sensitivity relatively as compared to the adjacent pair of electrodes. Accordingly, the inhomogeneous sensitivity distribution affects the accuracy of the reconstructed image. To increase the reconstruction performance in the ECVT, this study implemented a sectorized reconstruction approach based on sectoral sensitivity map correction. The basic idea is, instead of modifying the 3D sensitivity map in the whole region of interest, the sensitivity map is corrected only in the area needs to be focused. The corrected 3D sensitivity map was integrated in the iterative linear back projection (ILBP) to generate a volumetric image of an object using the ECVT. The results from simulation and experiment data showed an improvement in the quality of the reconstructed image.

     
    Keywords:

    Sectorized reconstruction, 3D Sensitivity matrix, Electrical Capacitance Volume Tomography