Image Reconstruction Algorithm Based On Wavelet Fusion For Electrical Capacitance Tomography

Electrical capacitance tomography (ECT) technology is a kind of processtomography (PT) technique that has been rapid development in the recent years.Because of low cost, non-intrusive detection, high safety and wide range ofapplication and so on, the ECT has become a research hotspot of processtomography technology. Since soft field effect and ill posed problem of theelectrical capacitance tomography system, the image reconstruction accuracy isgreatly limited. It has important significance and application value to carry outthe relevant theory and technology research.This thesis uses electrical capacitance tomography reconstruction problemas the research object, and is based on the analysis of the theory of imagereconstruction for electrical capacitance tomography. Launches the research inview of the problems of electrical capacitance tomography system is involved inimage reconstruction and quality evaluation of image reconstruction. The work ofthis thesis is as follows:Firstly, combined with the research background, we analyze the researchstatus of electrical capacitance tomography technology at present, anatomize thedevelopment trend of the technology, and also point out the combination ofwavelet fusion and ECT system.Secondly, based on analyzing the structure and working principle of ECTsystem, the finite element model to solve the forward problem is established andthe solving process of the sensitive field distribution function is given. Weanalyze several typical ECT reconstruction algorithms; obtain the evaluationindex for ECT image reconstruction.Again, the factors influencing the image fusion is analyzed, and the effect ofdifferent image fusion algorithm is discussed. These will provide the theory datasupport for the subsequent ECT image fusion. An image reconstruction algorithm based on wavelet fusion for electrical capacitance tomography is given. Thencomplete the experimental validation and results analysis.Finally, based on scalar wavelet and multi wavelet, using the maximum andregional gradient fusion rule, simulation experiment for low frequency part andhigh frequency part of the decomposed source images is given. Throughthe analysis of experimental results, multi wavelet is better than thatof scalar wavelet, can better improve the final accuracy of the ECT image.