Research On Image Reconstruction Algorithms Of Ultrasonic Computerized Tomography

With the development of modern industry, the accurate and online measurements for liquid-solid two-phase flow parameters in many industrial production departments including metallurgy, chemical, electric power industries are to have the vital significance. Ultrasonic computerized tomography receives wide attention in two phase flow measurement field because of its own advantages. But discrete phase medium has random distribution character and complicated physical properties in sound filed. So it requires that UCT system must have high accuracy for image reconstruction. The precision of UCT image reconstruction depends on image reconstruction algorithms’quality. Existing UCT reconstruction algorithms can’t satisfy the requirement for the resolution of liquid-solid two-phase flow. So our major objective is to find a reliable image reconstruction algorithm to structure high-precision UCT system.Firstly, according to the principle of ultrasonic attenuation imaging, the mathematical reconstruction model was structured. Tikhonov regularization method was introduced to solve this ill-posed image reconstruction problem. Based on classical Tikhonov regularization, we proposed a new regularization method--double parameters regularization algorithm which added the solution of overdetermined equations as a new restriction to solving underdetermined mathematical model. Then through a lot of simulation experiments, we summed up the method of selecting corresponding regularization parameters.Finally many simulation experiments were done to test the validity of double parameters regularization and traditional image reconstruction algorithm by simulation platform. The simulation results show that the double parameters regularization algorithm not only has a higher imaging accuracy but also make phase border clear and false depicts less. The algorithm expanded the regularization thought meaning and provided algorithm base for high precision UCT system. So it has important theoretical significance and practical value.