Research On Image Reconstruction Algorithms And Design Of Software For Electrical Impedance Tomography System

Electrical Impedance Tomography (EIT) is a new measurement technique developed in recent years. Due to the advantages of being non-radiant and non-intrusive, rapid response, simple structure and low cost, EIT has wide foreground for application in the field of clinical monitoring and industrial measurement.Although great progress has been made in EIT technique in the recent two decades, satisfactory images in terms of good quality and distinguishability haven't been achieved because the inverse problem is ill-conditioned. And since the researchers have paid more attention to the design of hardware and reconstruction algorithms, the development of the software system has lagged behind leading to deficient functions, unsatisfactory display effect and poor real-time performance. To solve the problems existed, thorough researches have been conducted mainly on image construction algorithms and design of software for EIT system in this thesis. The main work and results are as follows.1. Research on Image Reconstruction Algorithms for EIT Principles of algorithms commonly adopted for non-linear inverse problems have been summed up and applied to EIT image reconstruction. To solve ill-posed problem and poor resolution in EIT, a new image reconstruction algorithm based on total variation regularization is put forward. Compared with the conventional Tikhonov regularization, this new algorithm not only stabilizes the reconstruction, but also enhances the distinguishability of the reconstruction image in areas with discontinuous medium distribution, which possesses a good edge-preserving property. Both simulation and experimental results show that this algorithm has obvious advantages over other methods in resolution of the reconstruction image, and the real-time performance meets the requirements of application, too.2. Design of Software for EIT System A real-time EIT software system named EIT_SystemII has been designed based on 32-bits Windows platform and Visual C++ programming environment. It is composed of five function modules as follows: sensor model constructing & FEM meshing module; serial port communication module; algorithms application module; display module; data saving & replaying module. This software system is capable of constructing triangulation automatically according to the given sensor model. It realizes real-time data acquisition based on RS-232 communication and speeds up the...