Electrical capacitance volume tomography static imaging by non-optimized compressive sensing framework

Nur Afny Catur Andryani, Dodi Sudiana, Dadang Gunawan

Research output: Contribution to journalArticlepeer-review

10 Citations (Scopus)


Electrical capacitance volume tomography is a volumetric tomography technique that utilizes capacitance and fringing to capture behavior or perturbation in the sensing domain. One of the crucial issues in developing ECVT technology is the reconstruction algorithm. In practice, ILBP is most used due to its simplicity. However, it still presents elongation errors for certain dielectric contrasts. The high undersampling measurement of the ECVT imaging system, which is mathematically defined as an undetermined linear system, is one of the most challenging issues. Compressive sensing (CS) is a framework that enables the recovery of a sparse signal or a signal that can be represented as sparse in a certain domain, by having a lower dimension of measurement data compared to the Shanon-Nyquist theorem. Thus, mathematically, this framework is promising for solving an undetermined linear system such as the ECVT imaging system. This paper discusses the possibility of developing an ECVT imaging technique for static objects based on a CS framework. Based on the simulation results, Non-optimized CS does not completely succeed in providing better ECVT imaging quality. However, it does provide more localized imaging compared to ILBP. In addition, by having fewer requirements for the measurement data dimension, the CS framework is promising for reducing the number of required electrodes.

Original languageEnglish
Pages (from-to)243-260
Number of pages18
JournalJournal of ICT Research and Applications
Issue number3
Publication statusPublished - 2016


  • Compressive sensing framework
  • Electrical capacitance volume tomography
  • Imaging (reconstruction) algorithm
  • Static imaging
  • Tomography imaging


Dive into the research topics of 'Electrical capacitance volume tomography static imaging by non-optimized compressive sensing framework'. Together they form a unique fingerprint.

Cite this