基于压缩感知理论的电容层析成像算法研究

发布时间：2018-03-30 19:23

本文选题：电容层析成像　切入点：图像重建算法　出处：《华北电力大学》2017年硕士论文

【摘要】：电容层析成像(Electrical Capacitance Tomography,ECT)技术是自20世纪80年代发展起来的一种用于检测工业管道中多相流的过程层析成像技术。具有非辐射、非侵入、结构简单、成本低、安装携带方便、响应速度快等优点,现已成为国内外过程参数检测人员研究的焦点。电容层析成像图像重建技术其反问题求解过程存在的欠定性和病态性问题,始终成为制约图像重建精度进一步提高的瓶颈问题。2006年,Donoho等人提出了压缩感知(Compressed Sensing,CS)理论。该理论利用信号的稀疏性或可压缩性可以对信号进行少量采样并实现信号的精确重建。本文将压缩感知理论应用于电容层析成像图像重建过程中,以期解决电容层析成像欠定性问题,主要工作和结果如下:(1)根据ECT系统测量投影信号的特点,对投影信号在不同稀疏基空间中的稀疏效果进行研究。最终选取离散傅里叶变换基作为其稀疏基,具有较好的稀疏性;(2)深入研究压缩感知理论所要求的采样方式,对ECT系统固有采样方法进行改进,提出了两种随机采样方式,以减小稀疏基同观测矩阵之间的相关性;(3)对目前流行的四种基于压缩感知理论的信号重建算法进行深入研究,将其应用于电容层析成像图像重建过程中,并进行仿真实验。结果表明,将压缩感知理论应用于ECT图像重建中可以有效地提高重建图像质量。
[Abstract]:Electrical Capacitance tomography (ECT) is a process tomography technique developed since 1980s to detect multiphase flow in industrial pipelines. It has the advantages of non-radiation, non-invasion, simple structure, low cost and convenient installation and carrying. Because of its high response speed, it has become the focus of the researchers in the field of process parameter detection at home and abroad. The problem of ill-quality and ill-condition in the inverse problem solving process of the electrical capacitance tomography image reconstruction technology is discussed in this paper. Donoho et al put forward the theory of compressed sensing CSA in 2006. This theory can use the sparsity or compressibility of signal to sample a small amount of signal and realize the signal. In this paper, the theory of compression perception is applied to the reconstruction of electrical capacitance tomography image. The main work and results are as follows: 1) according to the characteristics of ECT system to measure projection signal, The sparse effect of projection signal in different sparse base space is studied. Finally, the discrete Fourier transform (DFT) basis is chosen as its sparse basis, which has good sparsity. The inherent sampling method of ECT system is improved and two random sampling methods are proposed to reduce the correlation between sparse basis and observation matrix. It is applied in the process of electrical capacitance tomography image reconstruction, and the simulation results show that the compression sensing theory can effectively improve the image quality of ECT image reconstruction.
【学位授予单位】：华北电力大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.41

【参考文献】