Image reconstruction in electrical capacitance tomography using multiple linear regression and regularization

A new reconstruction algorithm for electrical capacitance tomography (ECT) is proposed. The basic idea behind this algorithm is to model the forward problem by using multiple linear regression, then obtain a stabilized solution of the inverse problem by using a regularization technique. This algorithm is numerically simple and computationally fast because it involves only a single matrix-vector multiplication when reconstructing an image. Tests performed on experimental and simulation data indicate that this algorithm can provide images superior to those obtained with the linear back-projection (LBP) algorithm, with a similar reconstruction time.