Paper Info
Title
Joint 3d Alignment-Reconstruction Multi-Scale Approach For Cryo Electron Tomography
Abstract
3D volume reconstruction in cryo-electron tomography is possible by using Transmission Electron Microscope (TEM) images from different tilt angles. The misalignment of these images is one of the limits to the quality of the reconstructed object. There are many alignment techniques to deal with this problem. Their common feature is to correct the 2D geometric transformation in the projection images. Nevertheless, 3D geometric transformation can occur in the TEM acquisition including tilt angular uncertainty. In this paper, we proposed a new multi-scale approach based on a Conjugate Gradient optimization of a cost function between the 3D reconstructed and the projection images with the purpose to find all the 3D parameters of geometric transformation. Tests on synthetic and real data prove the accuracy of our geometric transformation estimation.
Year
DOI
Venue
2016
10.1109/ISBI.2016.7493460
2016 IEEE 13TH INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING (ISBI)
Keywords
Field
DocType
Electron tomography, 3D reconstruction, Geometric transformation uncertainty, Conjugate Gradient, Multi-scale
Conjugate gradient method,Computer vision,Electron tomography,Computer science,Optics,Tomography,Geometric transformation,Artificial intelligence,Volume reconstruction,Cryo-electron tomography,3D reconstruction
Conference
ISSN
Citations 
PageRank 
1945-7928
0
0.34
References 
Authors
4
5
Name
Order
Citations
PageRank
Hmida Rojbani132.16
Etienne Baudrier2448.72
Benoît Naegel320718.76
Loïc Mazo472.00
Atef Hamouda54012.57