Electrical Impedance Tomography reconstruction using ℓ1 norms for data and image terms
Electrical Impedance Tomography reconstruction using ℓ1 norms for data and image terms
复制标题
DOI：
10.1109/iembs.2008.4649764
复制
发表时间：
2008-10
期刊：
影响因子：
--
通讯作者：
Tao Dai;A. Adler
中科院分区：
文献类型：
--
作者：
Tao Dai;A. Adler
电阻抗断层扫描（EIT）通过对身体表面的电流模拟和电压测量来计算身体内的内部电导率分布。EIT的两个主要技术难点是空间分辨率低和对测量误差敏感。与传统的使用102范数的重建相比，使用101范数的图像重建允许解决这两个困难。数据残差项上的0.01范数降低了对测量误差的敏感性，而图像先验上的0.01范数降低了边缘模糊。本文提出并验证了一种用于EIT重建的一般滞后扩散型迭代方法，该方法可以根据数据残差和/或图像先验部分灵活地选择最小化。结果表明了该算法的灵活性和RMB 1解的优点。
Electrical Impedance Tomography (EIT) calculates the internal conductivity distribution within a body from current simulation and voltage measurements on the body surface. Two main technical difficulties of EIT are its low spatial resolution and sensitivity to measurement errors. Image reconstruction using ℓ1 norms allows addressing both difficulties, in comparison to traditional reconstruction using ℓ2 norms. A ℓ1 norm on the data residue term reduces the sensitivity to measurement errors, while the ℓ1 norm on the image prior reduces edge blurring. This paper proposes and tests a general lagged diffusivity type iterative method for EIT reconstructions ℓ1 and ℓ2 minimizations can be flexibly chosen on the data residue and/or image prior parts. Results show the flexibility of the algorithm and the merits of the ℓ1 solution.