The Three Dimensional Image Reconstruction Of Electrical Capacitance Tomography Based On Support Vector Machine

Electrical Capacitance Tomography(ECT)is a kind of process tomography technology developed earlier which has the advantages of non-invasion,non-contact and low cost.At present,the main problems of ECT technology focus on the sensitive field “soft field” in this technology,which leads to the low imaging accuracy and unsatisfactory speed in the process of image reconstruction.Therefore,image reconstruction has become a crucial link in ECT technology.Support Vector Machines(SVM)is a widely used algorithm in the field of machine learning.In this article,the mathematical principle behind the three-dimensional model of the three components of the overall structure of THE ECT system was analyzed in detail,and the capacitive sensor,data acquisition system and image reconstruction in the three-dimensional ECT system were described respectively.In this article,the ECT system of 24 electrodes with 8 electrodes in each layer of axial three-layer electrodes is studied and analyzed,and the simulation experiment of 24 electroplate capacitance detection sensor is carried out with ANSYS finite element simulation software.The model is subdivided into equal volume units.Through designed experiments to explore the different sensor structure parameters for the influence of ECT system,by comparing the experimental analysis of several different electrode incentives for the effects of the final image including single electrode and multielectrode incentives and put them as for carries on the experiment in a different environment.The differences in their adaptation to different environments were obtained.The image reconstruction algorithm used in this article is the support vector machine algorithm.Firstly,some basic principles of support vector machine are analyzed in detail,from the basic theory in statistical learning to how to solve the optimal classification hyperplane in specific applications under different circumstances.Finally,analyzes the weakness of the SVM algorithm in ECT image reconstruction and improved,through clustering algorithm will decrease the sample data set and the cluster center operators will be treated as raw data set for dimension reduction,improved the accuracy and efficiency of image reconstruction.