Open AccessElectrical capacitance tomography image reconstruction by improved orthogonal matching pursuit algorithm
Author(s) -
Yan Hua,
Wang Yan,
Wang Yi Fan,
Zhou Ying Gang
Publication year - 2020
Publication title -
iet science, measurement and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.418
H-Index - 49
eISSN - 1751-8830
pISSN - 1751-8822
DOI - 10.1049/iet-smt.2019.0255
Subject(s) - matching pursuit , iterative reconstruction , algorithm , compressed sensing , electrical capacitance tomography , tikhonov regularization , reconstruction algorithm , mathematics , energy (signal processing) , signal reconstruction , inverse problem , artificial intelligence , computer science , capacitance , signal processing , statistics , electrode , mathematical analysis , telecommunications , radar , chemistry
In order to improve the quality of reconstructed images in electrical capacitance tomography (ECT), the image reconstruction method based on compressed sensing for ECT is studied. First, the traditional discrete Fourier transform and discrete cosine transform are used as a sparsity basis to make the grey vectors of the typical two‐phase flow distributions sparse. The energy loss of the sparse signals under different sparsity degrees is calculated, and the effect of energy loss on the quality of reconstructed images is studied. Then, using the natural sparsity of the original signal, an improved orthogonal matching pursuit algorithm for ECT image reconstruction is proposed. There are two main improvements in the proposed algorithm. First, multiple columns instead of one column in each iteration are selected for improving the reconstruction speed. Second, a regularisation solution instead of the least‐squares solution is used for improving the adaptability to ill‐posed inverse problems. Simulation and experimental tests are carried out and the results show that the proposed method can effectively improve the reconstructed images quality, and on the whole, obtain better reconstruction results than the Landweber iteration algorithm, the Tikhonov regularisation algorithm, and the gradient projection for sparse reconstruction algorithm.