期刊名称：International Journal of Signal Processing, Image Processing and Pattern Recognition
印刷版ISSN：2005-4254
出版年度：2016
卷号：9
期号：9
页码：243-252
出版社：SERSC
摘要：Electrical capacitance tomography (ETC) is a kind of imaging technology which is based on the difference in the sensitivity distribution of capacitance in each medium. In under the background of the ETC system of 12 electrode capacitance sensor arrays an improved NSSN neural network algorithm will be applied to the reconstruction of the image of ECT system in this thesis, and making the ECT image reconstruction algorithm more stable and efficient. In training of the neural network algorithm with large scale, we adopt the method of dividing the sub network to improve it. Through the closed pipeline of gas and solid two-phaseflow monitoring data and using the improved neural network image algorithm for image reconstruction. The experimental results show that the improved method makes up for the deficiency of the slow speed of the large scale neural network operation.,simplifythe neural network structure, for electrical capacitance tomography provides a new way of thinking.
关键词：Electrical capacitance tomography; New NSSN network; Partition sub ;network; Image reconstruction
