Image reconstruction is a key problem for Electrical Resistance Tomography (ERT). Because of the soft-field nature and the ill-posed problem in solving inverse problem, traditional image reconstruction methods cannot achieve high accuracy and the process is usually time consuming. Since deep learning is good at mapping complicated nonlinear function, a deep learning method based on Convolutional Neural Network (CNN) is proposed for image reconstruction of ERT. To establish the database, 41122 samples were generated with numerical simulations. 10- fold cross validation was used to divide all samples into training set and validation set. The network structure was based on LeNet, and refined by applying dropout layer and moving average. After 346 training epochs, the Image Correlation Coefficient (ICC) on validation set was 0.95. When white Gaussian noise with a signal-to-noise ratio of 30, 40, and 50 were added to validation set, the ICC was 0.79, 0.89 and 0.93 respectively, which proved the anti-noise capability of the network. The reconstruction results on samples which have more inclusions, different conductivity and other shapes explained the network has good generalization ability. Furthermore, experimental data from a 16-electrode industrial ERT system was used to compare the accuracy of the proposed model with some typical reconstruction methods. Results show that the proposed CNN method has better reconstruction results than LBP, Tikhonov and Landweber.

As a non-invasive and radiation-free imaging modality, electrical impedance tomography (EIT) has attracted much attention in the last two decades and owns many industry and biomedical applications. However, due to the nonlinearity and ill-posedness of its inverse problem, the EIT images always suffer from low spatial resolution and are sensitive to the modelling errors. To achieve high resolution and modelling error robust EIT image, a two-stage deep learning (TSDL) method is proposed. The proposed method consists of a pre-reconstruction block and a convolutional neural network (CNN). The pre-reconstruction block learns the regularization pattern from the training data set and provides a rough reconstruction of the target. The CNN post-processes the pre-reconstruction result in a multi-level feature analysis strategy, and eliminates the modelling errors with prior information of the observation domain shape. The pre-reconstruction and CNN blocks are trained together by using a minimum square approach. To evaluate the performance of the TSDL method, the lung EIT problem was studied. The training data set is calculated from more than 100 thousand EIT simulation models generated from CT scans across 792 patients. Lung injury, measurement noise and model errors are randomly simulated during the model generation process. The trained TSDL model is evaluated with simulation testes, as well as the experimental tests from a laboratory setting. According to the results, the TSDL method could achieve high accuracy shape reconstructions and is robust against measurement noise and modelling errors.

Frequency-difference electrical impedance tomography (fdEIT) was originally developed to mitigate the systematic artifacts induced by modeling errors when a baseline dataset is unavailable. Instead of fine anatomical imaging, only coarse anomaly detection has been addressed in current fdEIT research mainly due to its low spatial resolution. On the other hand, there has been not enough study on fdEIT reconstruction algorithm as well. In this article, we propose an efficient and high-spatial-resolution algorithm for simultaneously reconstructing multiple fdEIT frames corresponding to inject currents with multiple frequencies. The electrical impedance tomography reconstruction problem is considered within a hierarchical Bayesian framework, where both intratask spatial clustering and intertask dependency are automatically learned and exploited in an unsupervised manner. The computation is accelerated by adopting a modified marginal likelihood maximization approach. Real-data experiments are conducted to verify the recovery performance of the proposed algorithm.

Multiphase flow is a commonly seen transient and complex dynamic system in many industrial processes. The phase fraction and velocity are two of the most important parameters for flow monitoring and measurement. Due to the advantages of simplicity in sensor structure, low fabrication costs and fast response, conductance sensors have received broad attentions in horizontal, vertical and inclined multiphase pipe flow measurement. A conductance sensor measures the multiphase mixture conductivity between two electrodes in contact with the fluid to determine the phase fraction. Combined with cross-correlation technique, the velocity can also be acquired. This paper presents a review on the basic sensing principle, different types of conductance sensors and their applications in flow monitoring, flow pattern identification, phase fraction determination and velocity measurement of multiphase flow. Finally, based on the conclusion of the disadvantages, advantages and limitations of this technique, the insight into trends for future development are given.

Monitoring multiphase flow process and measuring its parameters have been a research topic that received broad attentions for the past decades. Ultrasonic process tomography (UPT) obtains the distribution of the two-phase flow based on ultrasound propagation in different fluids, thus is valuable to the industrial monitoring and measurement. UPT can non-intrusively explore transient hydrodynamics of the multiphase flow, but th reconstruction quality of single modality is always dependent on phase distribution. In this paper, a dual-mode UPT that fuses the attenuation and time-of-flight (TOF) of ultrasound is presented. To implement this scheme, a digital dual-mode UPT system that works under both the transmission-mode and reflection-mode is designed. A 16-channel ultrasound data acquisition system is proposed, which is flexibly programmable and reconfigurable through an FPGA. The system includes a series of functional modules, and is based on the compact peripheral component interconnect (CPCI) bus for real-time data transmission in industrial standard. Static experiments show that the system can distinguish the interface of the different inclusions and preferably reconstruct the position and size of the inclusions through fusing the transmission and reflection reconstruction, the average Relative Error and Correlation Coefficient of the reconstructed images are 0.617 and 0.626 respectively, which show preferable and stable distinguishability of different inclusions.