3D Alignment of Projections in Electron Tomography
Tang, Qi (2016)
2016
Master's Degree Programme in Information Technology
Tieto- ja sähkötekniikan tiedekunta - Faculty of Computing and Electrical Engineering
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2016-01-13
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tty-201512161834
https://urn.fi/URN:NBN:fi:tty-201512161834
Tiivistelmä
The goal of this thesis is to analyze the effect of projection angle errors on reconstructing 3D electron tomography. Noise, missing wedge and miss alignment are three main problems in electron tomography. This thesis focuses on how miss alignment affects the 3D reconstruction relative to the noise and missing wedge effects. Fourier-based iterative method (FIRM) is the main reconstruction method in this work. Instead of projecting the volume and back-projecting the tilt series in the image domain, FIRM conducts those operations in Frequency domain. By using non-uniform fast Fourier transform, FIRM avoids interpolation problem. Besides, the conjugate gradient method is used as the iterative reconstruction process to find the optimal solution.
With the simulations, the impacts of noise, missing wedge and miss alignment are studied quantitatively. Missing wedge is the most influential factor among those three factors, the lack of enough information causes the reconstruction volume to be highly burred. Miss alignment has a similar effect as Gaussian noise. However, miss alignment also introduces artifacts in the reconstruction. Normalized mean-square error (NMSE) decreases and resolution increases when decreasing the level of miss alignment. After the level of 0.2-degree variance of projection angle error, the results does not show significant differences.
With the simulations, the impacts of noise, missing wedge and miss alignment are studied quantitatively. Missing wedge is the most influential factor among those three factors, the lack of enough information causes the reconstruction volume to be highly burred. Miss alignment has a similar effect as Gaussian noise. However, miss alignment also introduces artifacts in the reconstruction. Normalized mean-square error (NMSE) decreases and resolution increases when decreasing the level of miss alignment. After the level of 0.2-degree variance of projection angle error, the results does not show significant differences.