Abstract
Electrical capacitance tomography (ECT) is considered to be a competitive measurement method. The imaging objects in ECT measurements are often in a time-varying process, and exploiting the prior information related to the dynamic nature is important for reconstructing high-quality images. Different from existing reconstruction models, in this paper a new model that incorporates the spatial correlation of the pixels by introducing the radial basis function (RBF) method, the dynamic behaviors of a timevarying imaging object, and the ECT measurement information is proposed to formulate the dynamic imaging problem. An objective functional that exploits the spatial correlation of the pixels, the combinational regularizer of the first-order total variation (FOTV) and the second-order total variation (SOTV), the multi-scale regularization, the spatial constraint, and the temporal correlation is proposed to convert the ECT imaging task into an optimization problem. A split Bregman iteration (SBI) method based iteration scheme is developed for solving the proposed objective functional. Numerical simulation results validate the superiority of the proposed reconstruction method on the improvement of the imaging quality.
ECT-LSTM-RNN: An Electrical Capacitance Tomography Model-based Long Short-Term Memory Recurrent Neural Networks for Conductive Materials
