Fast Electromagnetic Tomography Image Reconstruction Algorithm Based on Dimension Reduction Technique
- PDF / 2,652,416 Bytes
- 20 Pages / 439.37 x 666.142 pts Page_size
- 21 Downloads / 180 Views
Fast Electromagnetic Tomography Image Reconstruction Algorithm Based on Dimension Reduction Technique Jiwei Huo1 · Ze Liu1 · Yadong Wang1 · Wei Yuan1 · Chengfei Wang1 Received: 27 May 2020 / Revised: 27 September 2020 / Accepted: 6 October 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Electromagnetic tomography systems that used in many industrial fields require high speed of image reconstruction. In this paper, a fast image reconstruction algorithm of electromagnetic tomography is proposed. For this method, firstly, the dimension of the sensitivity matrix is decreased through retention of principal component. Then, the distribution image is reconstructed by using traditional or improved reconstruction algorithm. This method can eliminate the redundant information of sensitivity matrix with little loss of feature information of sensitivity matrix. The simulation and experiment results show that compare with traditional image reconstruction algorithm (including iterative algorithm and single step algorithm), the proposed algorithm has significant advantage of imaging speed. The quality of reconstructed image that using traditional algorithm and reconstructed image that using proposed algorithm are almost same. Keywords EMT · Image reconstruction · Demension reduction · High-speed
1 Introduction Measurement technique of electromagnetic tomography (EMT) is based on the measurement of inductance by using detectors mounted around a circular region of interest and have advantages of low cost, non-invasiveness, non-contact and nonhazard [1–4], which has a wide variety of applications (e.g. coating inspection, monitoring metal producing processes, rail inspection, biomedical application and oil industry [5–12]). There are many factors are related to the performance of reconstructed image in an EMT system. The factors include frequency of excitation signal, image reconstruction algorithm, structure of detectors, et. al. Among these factors, reconstruction algorithm is a critical aspect. Over the past decade, there has been considerable interest in the * Ze Liu [email protected] 1
School of Electronic Information Engineering, Beijing Jiaotong University, Beijing, China
13
Vol.:(0123456789)
52
Page 2 of 20
Sensing and Imaging
(2020) 21:52
optimization of image reconstruction algorithm in order to improve performance of imaging so as to meet industrial demands. The image reconstruction including linear back-projection (LBP), Tikhonov regularization, Landweber iteration algorithm, Tikhonov iterative algorithm, Newton–Raphson iterative algorithm, conjugate gradient algorithm, et.al [13–17]. Because the advantage of effectiveness and convenience, the algorithm of LBP, Tikhonov regularization and Landweber iterative are widely used in EMT study [18]. In this paper, A method based on dimension reduction of sensitivity matrix is proposed to modify the reconstruction algorithm and increase the speed of imaging to feet the requirements of high-speed application condition, such as,
Data Loading...
