Modified sparse regularization for electrical impedance tomography
- School of Aeronautical Automation, Civil Aviation University of China, Tianjin 300300 (China)
- School of Electrical and Automation Engineering, Tianjin University, Tianjin 300072 (China)
- School of Electronics and Information Engineering, Tianjin Polytechnic University, Tianjin 300387 (China)
Electrical impedance tomography (EIT) aims to estimate the electrical properties at the interior of an object from current-voltage measurements on its boundary. It has been widely investigated due to its advantages of low cost, non-radiation, non-invasiveness, and high speed. Image reconstruction of EIT is a nonlinear and ill-posed inverse problem. Therefore, regularization techniques like Tikhonov regularization are used to solve the inverse problem. A sparse regularization based on L{sub 1} norm exhibits superiority in preserving boundary information at sharp changes or discontinuous areas in the image. However, the limitation of sparse regularization lies in the time consumption for solving the problem. In order to further improve the calculation speed of sparse regularization, a modified method based on separable approximation algorithm is proposed by using adaptive step-size and preconditioning technique. Both simulation and experimental results show the effectiveness of the proposed method in improving the image quality and real-time performance in the presence of different noise intensities and conductivity contrasts.
- OSTI ID:
- 22597092
- Journal Information:
- Review of Scientific Instruments, Vol. 87, Issue 3; Other Information: (c) 2016 AIP Publishing LLC; Country of input: International Atomic Energy Agency (IAEA); ISSN 0034-6748
- Country of Publication:
- United States
- Language:
- English
Similar Records
Variable-Precision Arithmetic for Solving Inverse Problems of Electrical Impedance Tomography
Inverse transport problem solvers based on regularized and compressive sensing techniques