scholarly journals Symmetric and Transitive Registration of Image Sequences

2008 ◽  
Vol 2008 ◽  
pp. 1-9 ◽  
Author(s):  
Oskar Škrinjar ◽  
Arnaud Bistoquet ◽  
Hemant Tagare
Keyword(s):  
Image Registration ◽  
Motion Tracking ◽  
Magnetic Resonance Images ◽  
Image Sequences ◽  
Pixel Size ◽  
Machine Precision ◽  
Registration Accuracy ◽  
Nonrigid Image Registration ◽  
Registration Algorithm ◽  
Image Type

This paper presents a method for constructing symmetric and transitive algorithms for registration of image sequences from image registration algorithms that do not have these two properties. The method is applicable to both rigid and nonrigid registration and it can be used with linear or periodic image sequences. The symmetry and transitivity properties are satisfied exactly (up to the machine precision), that is, they always hold regardless of the image type, quality, and the registration algorithm as long as the computed transformations are invertable. These two properties are especially important in motion tracking applications since physically incorrect deformations might be obtained if the registration algorithm is not symmetric and transitive. The method was tested on two sequences of cardiac magnetic resonance images using two different nonrigid image registration algorithms. It was demonstrated that the transitivity and symmetry errors of the symmetric and transitive modification of the algorithms could be made arbitrary small when the computed transformations are invertable, whereas the corresponding errors for the nonmodified algorithms were on the order of the pixel size. Furthermore, the symmetric and transitive modification of the algorithms had higher registration accuracy than the nonmodified algorithms for both image sequences.

Non-Rigid Image Registration based on Parameterized Surfaces: Application to 3D Cardiac Motion Image Analysis

2021 ◽  
pp. 2250028
Author(s):  
Said Khalid Shah
Keyword(s):  
Image Registration ◽  
Motion Tracking ◽  
Salient Feature ◽  
Ct Images ◽  
Heart Beat ◽  
Cardiac Motion ◽  
Free Form ◽  
3D Ct ◽  
Registration Accuracy ◽  
Whole Heart

This paper describes the Fast Radial Basis Function (RBF) method for cardiac motion tracking in 3D CT using non-rigid medical image registration based on parameterized (regular) surfaces. The technique is a point-based registration evaluation algorithm which does register 3D MR or CT images in real time. We first extract the surface of the whole heart 3D CT and its contrast enhanced part (left ventricle (LV) blood cavity) of each dataset with a semiautomatic contouring and a fully-automatic triangulation method followed by a global surface parameterization and optimization algorithm. In second step, a set of registration experiments are run to calculate the deformation field at various phases of cardiac motion or cycle from CT images, which results into significant deformation during each phase of a cycle. The surface points of the whole heart and LV are used to register the source systole image to various diastole target images taken at different phases during a heart beat. Our registration accuracy improves with the increase in number of salient feature points (i.e. optimized parameterized surfaces) and it has no effect on the speed of the algorithm (i.e. still less than a second). The results show that the target registration error is less than 3[Formula: see text]mm (2.53) and the performance of the Fast RBF algorithm is less than a second using a whole heart CT dataset of a single patient taken over the course of the entire cardiac cycle. At the end, the results for recovery (or analysis) of bigger deformation in heart CT images using the Fast RBF algorithm is compared to the state-of-the-art Free Form Deformation (FFD) registration technique. It is proved that the Fast RBF method is performing better in speed and slightly less accurate than the FFD (when measured in terms of NMI) due to iterative nature of the latter.

Effect of Measure and Interpolation Methods for Intensity-Based Medical Image Registration

2013 ◽  
Vol 647 ◽  
pp. 612-617
Author(s):  
Guo Dong Zhang ◽  
Xiao Hu Xue ◽  
Wei Guo
Keyword(s):  
Image Registration ◽  
Mutual Information ◽  
Medical Image ◽  
Medical Image Registration ◽  
Registration Accuracy ◽  
Partial Volume ◽  
Interpolation Methods ◽  
Registration Algorithm ◽  
Image Registration Algorithm ◽  
Local Extreme

The local extreme is main reason to hamper the optimization process and influence the registration accuracy in medical image registration algorithm. In general, the accuracy of image registration based on mutual information is afforded by interpolation methods. In this paper, we analyze the effect of the measure and interpolation methods for medical image registration and present a medical image registration algorithm using mutual strictly concave function measure and partial volume (PV) interpolation methods. The experiment results show that for images with low local correlation the algorithm has the ability to reduce the local extreme, the registration accuracy is improved, and the algorithm expended less time than mutual information based registration method with partial volume (PV) or generalized partial volume estimation (GPVE).

scholarly journals Transitively Consistent and Unbiased Multi-Image Registration using Numerically Stable Transformation Synchronisation

2015 ◽  
Author(s):  
Florian Bernard ◽  
Johan Thunberg ◽  
Andreas Husch ◽  
Luis Salamanca ◽  
Peter Gemmar ◽  
...  
Keyword(s):  
Image Registration ◽  
Medical Images ◽  
Desirable Property ◽  
Stable Transformation ◽  
Affine Transformations ◽  
Registration Accuracy ◽  
Brain Images ◽  
Registration Algorithm ◽  
Image Registration Algorithm ◽  
Affine Image

Transitive consistency of pairwise transformations is a desirable property of groupwise image registration procedures. The transformation synchronisation method (Bernard et al., 2015) is able to retrieve transitively consistent pairwise transformations from pairwise transformations that are initially not transitively consistent. In the present paper, we present a numerically stable implementation of the transformation synchronisation method for affine transformations, which can deal with very large translations, such as those occurring in medical images where the coordinate origins may be far away from each other. By using this method in conjunction with any pairwise (affine) image registration algorithm, a transitively consistent and unbiased groupwise image registration can be achieved. Experiments involving the average template generation from 3D brain images demonstrate that the method is more robust with respect to outliers and achieves higher registration accuracy compared to reference-based registration.

Registration for SAR and Optical Image via Cross Cumulative Residual Entropy and Ratio Operator

2012 ◽  
Vol 452-453 ◽  
pp. 954-958 ◽  
Author(s):  
Yong Mei Zhang ◽  
Jie Qiong Li
Keyword(s):  
Image Registration ◽  
Optical Image ◽  
Residual Entropy ◽  
Sar Image ◽  
Registration Accuracy ◽  
Sar Images ◽  
Registration Method ◽  
Registration Algorithm ◽  
Cumulative Residual Entropy ◽  
Cumulative Residual

Optical and SAR images have different imaging modes and pixel expression formats, which caused much difficult for image registration. The paper proposes a registration algorithm combined Ratio gradient and Cross Cumulative Residual Entropy (CCRE) aimed at solving the problem of SAR image speckles. Compared with other traditional methods, the experiment results show that the CCRE combined with Ratio operator registration method performs satisfactorily in SAR and optical image registration and provides a significant improvement on the registration accuracy over the other algorithm

Sign in / Sign up

Export Citation Format

Share Document