Image Registration Problem Research Articles

Elastic registration of prostate MR images based on estimation of deformation states.

Magnetic resonance imaging (MRI) is being used increasingly for image-guided targeted biopsy and focal therapy of prostate cancer. In this paper, a combined rigid and deformable registration technique is proposed to register pre-treatment diagnostic 3T magnetic resonance (MR) images of the prostate, with the identified target tumor(s), to intra-treatment 1.5T MR images. The pre-treatment T2-weighted MR images were acquired with patients in a supine position using an endorectal coil in a 3T scanner, while the intra-treatment T2-weighted MR images were acquired in a 1.5T scanner before insertion of the needle with patients in the semi-lithotomy position. Both the rigid and deformable registration algorithms employ an intensity-based distance metric defined based on the modality independent neighborhood descriptors (MIND) between images. The optimization routine for estimating the rigid transformation parameters is initialized using four pairs of manually selected approximate corresponding points on the boundaries of the prostate. In this paper, the problem of deformable image registration is approached from the perspective of state estimation for dynamical systems. The registration algorithm employs a rather generic dynamic linear elastic model of the tissue deformation discretized by the finite element method (FEM). We use the model in a classical state estimation framework to estimate the deformation of the prostate based on the distance metric between pre- and intra-treatment images. Our deformable registration results using 17 sets of prostate MR images showed that the proposed method yielded a target registration error (TRE) of 1.87 ± 0.94 mm,2.03 ± 0.94 mm, and 1.70 ± 0.93 mm for the whole gland (WG), central gland (CG), and peripheral zone (PZ), respectively, using 76 manually-identified fiducial points. This was an improvement over the 2.67 ± 1.31 mm, 2.95 ± 1.43 mm, and 2.34 ± 1.11 mm, respectively for the WG, CG, and PZ after rigid registration alone. Dice similarity coefficients (DSC) in the WG, CG and PZ were 88.2 ± 5.3, 85.6 ± 7.6 and 68.7 ± 6.9 percent, respectively. Furthermore, the mean absolute distances (MAD) between surfaces was 1.26 ± 0.56 mm and 1.27 ± 0.55 mm in the WG and CG, after deformable registration. These results indicate that the proposed registration technique has sufficient accuracy for localizing prostate tumors in MRI-guided targeted biopsy or focal therapy of clinically localized prostate cancer.

An Inexact Newton-Krylov Algorithm for Constrained Diffeomorphic Image Registration.

We propose numerical algorithms for solving large deformation diffeomorphic image registration problems. We formulate the nonrigid image registration problem as a problem of optimal control. This leads to an infinite-dimensional partial differential equation (PDE) constrained optimization problem. The PDE constraint consists, in its simplest form, of a hyperbolic transport equation for the evolution of the image intensity. The control variable is the velocity field. Tikhonov regularization on the control ensures well-posedness. We consider standard smoothness regularization based on H1- or H2-seminorms. We augment this regularization scheme with a constraint on the divergence of the velocity field (control variable) rendering the deformation incompressible (Stokes regularization scheme) and thus ensuring that the determinant of the deformation gradient is equal to one, up to the numerical error. We use a Fourier pseudospectral discretization in space and a Chebyshev pseudospectral discretization in time. The latter allows us to reduce the number of unknowns and enables the time-adaptive inversion for nonstationary velocity fields. We use a preconditioned, globalized, matrix-free, inexact Newton-Krylov method for numerical optimization. A parameter continuation is designed to estimate an optimal regularization parameter. Regularity is ensured by controlling the geometric properties of the deformation field. Overall, we arrive at a black-box solver that exploits computational tools that are precisely tailored for solving the optimality system. We study spectral properties of the Hessian, grid convergence, numerical accuracy, computational efficiency, and deformation regularity of our scheme. We compare the designed Newton-Krylov methods with a globalized Picard method (preconditioned gradient descent). We study the influence of a varying number of unknowns in time. The reported results demonstrate excellent numerical accuracy, guaranteed local deformation regularity, and computational efficiency with an optional control on local mass conservation. The Newton-Krylov methods clearly outperform the Picard method if high accuracy of the inversion is required. Our method provides equally good results for stationary and nonstationary velocity fields for two-image registration problems.

Continuous piecewise affine transformation for image registration

The image registration problem involves determining a geometric transformation to properly align images of interest. This paper proposes a transformation approach called Continuous Piecewise Affine Transformation (CPAT) to model the geometric distortion in images. The associated design methodology for the 2D registration problem is also discussed. Registration on CPAT has two advantages: (1) The optimal transformation has a closed form analytical solution; and (2) the transformation is invertible and transitive. Because of these advantages, CPAT could be used for real-time registration and progressive registration, where the registration speed and transitivity property are important for the successful completion of the tasks. The results of experiments demonstrate the efficacy of CPAT in resolving the image registration problem.

Motion Compensated Abdominal Diffusion Weighted MRI by Simultaneous Image Registration and Model Estimation (SIR-ME).

Non-invasive characterization of water molecule's mobility variations by quantitative analysis of diffusion-weighted MRI (DW-MRI) signal decay in the abdomen has the potential to serve as a biomarker in gastrointestinal and oncological applications. Accurate and reproducible estimation of the signal decay model parameters is challenging due to the presence of respiratory, cardiac, and peristalsis motion. Independent registration of each b-value image to the b-value=0 s/mm(2) image prior to parameter estimation might be sub-optimal because of the low SNR and contrast difference between images of varying b-value. In this work, we introduce a motion-compensated parameter estimation framework that simultaneously solves image registration and model estimation (SIR-ME) problems by utilizing the interdependence of acquired volumes along the diffusion weighting dimension. We evaluated the improvement in model parameters estimation accuracy using 16 in-vivo DW-MRI data sets of Crohn's disease patients by comparing parameter estimates obtained using the SIR-ME model to the parameter estimates obtained by fitting the signal decay model to the acquired DW-MRI images. The proposed SIR-ME model reduced the average root-mean-square error between the observed signal and the fitted model by more than 50%. Moreover, the SIR-ME model estimates discriminate between normal and abnormal bowel loops better than the standard parameter estimates.

A Novel Method To Collect Kinect-Comparable RGB-D Outdoor Image

A RGB-D based visual computing research has received increasing attention among researchers, such as pose estimation, segmentation, people detection, object labelling, and so on. However, almost all the existed RGB-D based researches are using indoor RGB-D datasets. The main reason is that there’s no available sensors for collecting outdoor RGB-D images. In this paper, we combine a ToF camera with a RGB camera and solve the image registration problem based on the pin-hole camera model. The initial experiments show that our method can perform as well as Kinect indoors but surpass Kinect outdoors.

RMI based non-rigid image registration using BF-QPSO optimization and P-spline

Registration is an image processing technique which aligns two or more images spatially and produces an informative image. Regional mutual information is an entropy based similarity measure that is invariant to the overlapped regions of the two images to be registered. But, it is difficult for images having deformation due to organ motion with a non-convex and irregular shape. In this paper, nonrigid image registration problem is addressed for multimodal medical images where P-spline interpolation method is adopted for modeling the transformation parameters. Intensity based registration requires optimization of similarity metric. As function of these parameters is complex, local optimization techniques frequently fail, which necessitates global optimization methods. Quantum-behaved particle swarm optimization (QPSO) requires fewer parameters to control and outperforms particle swarm optimization (PSO) in search ability. Bacterial Foraging Algorithm (BFA) is a non gradient global optimization technique where E. coli bacteria depends on random search direction during chemotaxis, which delays the rate of convergence. Here, a new algorithm is proposed hybridizing the notion of QPSO and BFA for tradeoff between local and global search. P-spline interpolated image registration is tested with 10 sets of MR images and some other medical images using RMI similarity index optimized by BF-QPSO. The proposed technique is compared with BFA, QPSO and hybrid BF-PSO algorithms.

Non-rigid registration of medical images based on estimation of deformation states

A unified framework for automatic non-rigid 3D–3D and 3D–2D registration of medical images with static and dynamic deformations is proposed in this paper. The problem of non-rigid image registration is approached as a classical state estimation problem using a generic deformation model for the soft tissue. The registration technique employs a dynamic linear elastic continuum mechanics model of the tissue deformation, which is discretized using the finite element method. In the proposed method, the registration is achieved through a Kalman-like filtering process, which incorporates information from the deformation model and a vector of observation prediction errors computed from an intensity-based similarity/distance metric between images. With this formulation, single and multiple-modality, 3D–3D and 3D–2D image registration problems can all be treated within the same framework. The performance of the proposed registration technique was evaluated in a number of different registration scenarios. First, 3D magnetic resonance (MR) images of uncompressed and compressed breast tissue were co-registered. 3D MR images of the uncompressed breast tissue were also registered to a sequence of simulated 2D interventional MR images of the compressed breast. Finally, the registration algorithm was employed to dynamically track a target sub-volume inside the breast tissue during the process of the biopsy needle insertion based on registering pre-insertion 3D MR images to a sequence of real-time simulated 2D interventional MR images. Registration results indicate that the proposed method can be effectively employed for the registration of medical images in image-guided procedures, such as breast biopsy in which the tissue undergoes static and dynamic deformations.

A comparative study on the application of advanced bacterial foraging models to image registration

New swarm intelligence approaches as the Bacterial Foraging Optimization Algorithm (BFOA) have recently awakened a growing interest in the evolutionary computation community. This fact is due to the promising results obtained by different variants of the latter optimization method in many real-world applications. In this work we aim to take a step ahead in the development of the discipline by introducing a large amount of BFOA variants resulting from the combination of some advanced design decisions applied on Dasgupta et al.’s self-adaptive version. Our goal is thus to achieve an improved understanding of the good properties that the BFOA approach has shown in previous contributions. We will perform an extensive experimental study considering a plethora of algorithmic variants to solve a real-world problem, image registration, a well-known and complex task in computer vision. In particular, more than fifty variants are proposed and tested tackling pair-wise image registration problem instances from two different domains, namely range image registration for 3D model reconstruction and 3D medical image registration. The reported results reveal that BFOA is a versatile approach able to provide very competitive outcomes to face challenging real-world image registration problems when compared to the state-of-the-art evolutionary approaches in the field.

A new affine invariant feature extraction method for SAR image registration

To cope with the problem of automatic registration of synthetic aperture radar (SAR) images, a novel affine invariant feature extraction method is proposed. The proposed method adapts the salient image disks (SIDs) extraction method to accurately localize affine invariant features in SAR images. The contributions mainly include three aspects. First, a multi-scale isotropic matched filter (MIMF) operator is modified to detect salient features. Then, a multi-resolution search strategy is adopted for eliminating unstable features, and affine invariance is achieved by an affine adaptation process. Finally, a point-set registration method based on Coherent Point Drift (CPD) is applied to the SAR image registration problem. Experiment results show that the proposed method has great improvement compared with the original method.

Improved image registration by sparse patch-based deformation estimation

Despite intensive efforts for decades, deformable image registration is still a challenging problem due to the potential large anatomical differences across individual images, which limits the registration performance. Fortunately, this issue could be alleviated if a good initial deformation can be provided for the two images under registration, which are often termed as the moving subject and the fixed template, respectively. In this work, we present a novel patch-based initial deformation prediction framework for improving the performance of existing registration algorithms. Our main idea is to estimate the initial deformation between subject and template in a patch-wise fashion by using the sparse representation technique. We argue that two image patches should follow the same deformation toward the template image if their patch-wise appearance patterns are similar. To this end, our framework consists of two stages, i.e., the training stage and the application stage. In the training stage, we register all training images to the pre-selected template, such that the deformation of each training image with respect to the template is known. In the application stage, we apply the following four steps to efficiently calculate the initial deformation field for the new test subject: (1) We pick a small number of key points in the distinctive regions of the test subject; (2) for each key point, we extract a local patch and form a coupled appearance-deformation dictionary from training images where each dictionary atom consists of the image intensity patch as well as their respective local deformations; (3) a small set of training image patches in the coupled dictionary are selected to represent the image patch of each subject key point by sparse representation. Then, we can predict the initial deformation for each subject key point by propagating the pre-estimated deformations on the selected training patches with the same sparse representation coefficients; and (4) we employ thin-plate splines (TPS) to interpolate a dense initial deformation field by considering all key points as the control points. Thus, the conventional image registration problem becomes much easier in the sense that we only need to compute the remaining small deformation for completing the registration of the subject to the template. Experimental results on both simulated and real data show that the registration performance can be significantly improved after integrating our patch-based deformation prediction framework into the existing registration algorithms.

Robust Registration by Rank Minimization for Multiangle Hyper/Multispectral Remotely Sensed Imagery

Multiangle images are acquired over roughly the same earth surface from different angles, and accurate image registration is a key prerequisite for their application. In this paper, we propose a robust registration method by rank minimization (RRRM) for multiangle hyper/multispectral remotely sensed imagery (MA-HSI-MSI). First, the low-rank structure of the MA-HSI-MSI is exploited and utilized as the registration constraint, thus recasting the image registration problem as searching for an optimal set of transformations, such that the matrix of the transformed images can reach its minimum rank. Second, a patch-based registration scheme is adopted to solve the problem of inconsistent geometric distortion over the entire image, taking the homography model as the local transformation. An iterative convex optimization algorithm is then used to solve the rank minimization-based image registration model for each image patch. Finally, all the transformed patches are used to synthesize the final registration image. The experimental results demonstrate that the proposed low-rank registration method works effectively for CHRIS/Proba imagery and WorldView-2 imagery.

Random sampling algorithm in the image registration problem

Image registration is one of the basic problems of computer vision. It arises in optical flow estimation, stereo vision, and tracking problems. One of the classical approaches proposed by B. Lucas and T. Kanade is based on optimization of some cost function. In this article image registration algorithm based on Lucas–Kanade approach is proposed (random sampling algorithm). It shows high performance results.

Computing global minimizers to a constrained B‐spline image registration problem from optimal l1 perturbations to block match data

Block matching is a well-known strategy for estimating corresponding voxel locations between a pair of images according to an image similarity metric. Though robust to issues such as image noise and large magnitude voxel displacements, the estimated point matches are not guaranteed to be spatially accurate. However, the underlying optimization problem solved by the block matching procedure is similar in structure to the class of optimization problem associated with B-spline based registration methods. By exploiting this relationship, the authors derive a numerical method for computing a global minimizer to a constrained B-spline registration problem that incorporates the robustness of block matching with the global smoothness properties inherent to B-spline parameterization. The method reformulates the traditional B-spline registration problem as a basis pursuit problem describing the minimall1-perturbation to block match pairs required to produce a B-spline fitting error within a given tolerance. The sparsity pattern of the optimal perturbation then defines a voxel point cloud subset on which the B-spline fit is a global minimizer to a constrained variant of the B-spline registration problem. As opposed to traditional B-spline algorithms, the optimization step involving the actual image data is addressed by block matching. The performance of the method is measured in terms of spatial accuracy using ten inhale/exhale thoracic CT image pairs (available for download atwww.dir-lab.com) obtained from the COPDgene dataset and corresponding sets of expert-determined landmark point pairs. The results of the validation procedure demonstrate that the method can achieve a high spatial accuracy on a significantly complex image set. The proposed methodology is demonstrated to achieve a high spatial accuracy and is generalizable in that in can employ any displacement field parameterization described as a least squares fit to block match generated estimates. Thus, the framework allows for a wide range of image similarity block match metric and physical modeling combinations.

A way to predict parameters of image registration by estimating inter-frame deformation of local fragments

For the problem of image registration (binding), we examine whether it is possible to increase the prediction accuracy of binding parameters between two centers of local image fragments for which binding parameters have been determined. We study how to use linear and cubic polynomials, a Bezier curve, a bi-arc curve, and a rational Bezier curve for this purpose.

Higher-order spatial accuracy in diffeomorphic image registration

Abstract: We discretize a cost functional for image registration problems by deriving Taylor expansions for the matching term. Minima of the discretized cost functionals can be computed with no spatial discretization error, and the optimal solutions are equivalent to minimal energy curves in the space of k-jets. We show that the solutions convergence to optimal solutions of the original cost functional as the number of particles increases with a convergence rate of O(hd+k) where h is a resolution parameter. The effect of this approach over traditional particle methods is illustrated on synthetic examples and real images.

An Orthogonal Learning Differential Evolution Algorithm for Remote Sensing Image Registration

We introduce an area-based method for remote sensing image registration. We use orthogonal learning differential evolution algorithm to optimize the similarity metric between the reference image and the target image. Many local and global methods have been used to achieve the optimal similarity metric in the last few years. Because remote sensing images are usually influenced by large distortions and high noise, local methods will fail in some cases. For this reason, global methods are often required. The orthogonal learning (OL) strategy is efficient when searching in complex problem spaces. In addition, it can discover more useful information via orthogonal experimental design (OED). Differential evolution (DE) is a heuristic algorithm. It has shown to be efficient in solving the remote sensing image registration problem. So orthogonal learning differential evolution algorithm (OLDE) is efficient for many optimization problems. The OLDE method uses the OL strategy to guide the DE algorithm to discover more useful information. Experiments show that the OLDE method is more robust and efficient for registering remote sensing images.

Nonrigid Image Registration Problem Using Fluid Dynamics and Mutual Information

Image registration is one of the fundamental tasks within image processing. It has wide applications in the fields of medical imaging, computer vision, statistical modeling etc. It is required when one wants to combine valuable statistical information from multiple images, possibly acquired using different modalities, at different time points or from different subjects, or to compare or integrate the data obtained from same or different measures. The subject of this article is nonrigid image registration. In particular, the main focus is on the application of fluid dynamics and mutual information (MI), and the comparison of two different similarity measures, i.e. mutual information and sum of squared differences (SSD). Numerical experiments show that fluid registration using SSD is ideal for mono-modal image registration, while fluid registration using MI does a better job in multi-modal image registration.

A Methodology of Three-Dimensional Medical Image Registration Based on Conformal Geometric Invariant

A three-dimensional multimodality medical image registration method using geometric invariant based on conformal geometric algebra (CGA) theory is put forward for responding to challenges resulting from many free degrees and computational burdens with 3D medical image registration problems. The mathematical model and calculation method of dual-vector projection invariant are established using the distribution characteristics of point cloud data and the point-to-plane distance-based measurement in CGA space. The translation operator and geometric rotation operator during registration operation are built in Clifford algebra (CA) space. The conformal geometrical algebra is used to realize the registration of 3D CT/MR-PD medical image data based on the dual vector geometric invariant. The registration experiment results indicate that the methodology proposed in this paper is of stronger commonality, less computation burden, shorter time consumption, and intuitive geometric meaning. Both subjective evaluation and objective indicators show that the methodology proposed here is of high registration accuracy and suitable for 3D medical image registration.

Robust Registration of Remote Sensing Image Based on SURF and KCCA

The Speeded Up Robust Features (SURF) description’s success for computer vision applications makes it an attractive solution for image registration problem. For remote sensing images, SURF feature matching is often impacted by similar structures. To overcome the mentioned problem, we propose to use spatial relationship along with the SURF descriptor for remote sensing image registration. Firstly, a putative set of correspondences is obtained based on distances between SURF feature descriptors. Secondly, the spatial relationship of matched features is accomplished based on Kernel Canonical Correlation Analysis (KCCA), and then an influence function is established by the spatial relationship to figure out false matches. Introduction of spatial relationship to the SURF descriptors not only reduces the number of false matches but also help to keep the number of correct matches. Experimental results show an overall significant reduction of the mismatches while maintaining a high rate of correct matches.

2D-3D Medical Image Registration Based on Ant Colony Algorithm

Object: To optimize the rigidity registration algorithm between X-ray fluoroscopy and CT, and improve the accuracy of registration. Method: By changing the transmission parameters of the ray tracing, it can obtain the original DRR images and the float DRR image for registration. In trials, it uses ant colony algorithm as the optimized search strategy and Mutual information as the similarity measure. Result: ant colony algorithm and the improved ant colony algorithm compared to the classic Powell algorithm to improve the accuracy of registration about 10% and 20%, achieved good results. Conclusion: Ant Colony Algorithm as optimization search strategy can effectively solve the local minima problem in 2D-3D medical image registration, and further improve the accuracy of registration.