Higher-Dimensional Affine Registration and Vision Applications

A fully automated application was developed and used for the registration of T1-weighted magnetic resonance images (MRIs) for Alzheimer patients. Two methods for image registration were implemented and compared: affine and nonlinear registration. Nonlinear registration uses continuum-mechanics-based elastic deformation. The affine registration algorithm is linear and is generated by an amplitude-modulated phase-only filter. The nonlinear registration method uses an elastic transformation generated by Navier-Stokes continuum-mechanics models. The validation method to quantitatively compare the performance of the affine and nonlinear registration algorithms uses root-mean-square error and three-dimensional volume rendering.

The traditional affine iterative closest point (ICP) algorithm is fast and accurate for affine registration between two point sets, but it is easy to fall into local minimum. This paper proposes a robust Affine ICP algorithm based on corner points. First, an objective function is established under the guidance of corner points, where the corner points as the shape control point guides the affine registration of 2D point sets. Then, at each step of the algorithm, the affine transformation obtained by the last iterative step is used to establish the correspondence of these two point sets. Next, the new affine transformation is solved by using the objective function under the guidance of the corner points. Experimental results demonstrate that the robustness and convergence of our algorithm are greatly improved compared with the traditional affine registration algorithm.

Accurate registration is a key step for precise and real-time measurement of appearance phenotype using the forward intersection method. For example, in order to reconstruct the details of the plant organs (leaf, fruit, stem, etc.), precisely matched image points from varied object points should be of high density in the image region of a target organ in each image. However, this task is not achieved stably with relevant methods in current literature. In this paper, the multiple basic parameters for affine registration are obtained by estimating the distance reference value, obtaining the camera attitude from the inertial module, calibrating the internal parameters of the camera, calibrating the structure parameters of the vision system, aberration modelling, and other methods. The basic parameters are used to express the vectors of object points to be measured, the optical vectors of the image points to be matched and the scale factors of the optical projection in varied camera coordinate systems. In the registration process, these vectors are affine operated to map the first target image point to each image via the strategy of mapping. The system error of the approximately matched pixel is eliminated in the micro-region of error centred around the approximately matched pixel. Research shows that this affine registration algorithm is able to obtain a large number of highdensity and accurate matched image points that correspond to different target points in the leaf region of each image. This is suitable for forward intersection measurement of the plant appearance phenotype details with a real-time vision system based on our results. The proposed registration method is highly robust, efficient and precise; and requires relatively little computation resources.

End-to-end affine registration framework for histopathological images with weak annotations

A procedure to improve the convergence rate for affine registration methods of medical brain images when the images differ greatly from the template is presented. The methodology is based on a histogram matching of the source images with respect to the reference brain template before proceeding with the affine registration. The preprocessed source brain images are spatially normalized to a template using a general affine model with 12 parameters. A sum of squared differences between the source images and the template is considered as objective function, and a Gauss-Newton optimization algorithm is used to find the minimum of the cost function. Using histogram equalization as a preprocessing step improves the convergence rate in the affine registration algorithm of brain images as we show in this work using SPECT and PET brain images.

Affine iterative closest point algorithm for point set registration

Quantitative first-pass magnetic resonance (MR) imaging studies assist in characterizing the severity of ischemic heart disease. Cardiac motion due to a patient's difficulty in maintaining the required breath hold is often severe, making registration necessary. In this paper we present a novel automatic affine registration algorithm. The algorithm utilizes an affine active contour to determine affine registration errors in first-pass MR studies. We compare our results to the original unregistered studies and a normalized cross-correlation registration method

Point set registration is significant for many applications such as recognition and reconstruction problems in computer science, and localization and mapping problems in robotic science. Traditional iterative closest point (ICP) algorithm is fast, but it is only suitable for registration of rigid motions. Traditional affine ICP algorithm is fast enough and can match the shapes non-rigidly transformed, but it is not robust for noises and outliers. In this study, we propose a new affine ICP variant using correntropy criterion and point-to-line metric. Correntropy is a similarity measure between two random variables and it has outlier rejection property. By maximizing the objective function defined, the registration performance of affine ICP is increased. The method proposed is also find transformation as fast as traditional affine ICP algorithm. Experimental studies on 2D shapes show that our method is quite good in affine registration with noise and outliers in terms of accuracy and speed. The results are compared with state-of-the-art methods.

Fingerprint matching is the most important part in the field of fingerprint recognition. In this paper, a novel fingerprint matching algorithm based on the probabilistic graphical model and 3-tree model is proposed. First, minutiae matching problems are considered as a special point-set matching. Fingerprint minutiae are viewed as random variables. Each minutia pairs have some probability to be matched. Second, an algorithm is proposed to generate the graphical model and choose "signal points", which dynamically have corresponding points in other point set. We choose three base minutiae pairs as signal pairs. Third, the model is converted into a Junction Tree. A 3-tree model is built and the potentials of other minutiae pairs are calculated through Junction Tree (J.T.) algorithm. Then we translate the matching problem into the best matching problem of a weighted bipartite graph. Finally, the number of common matching pairs can be got through maximum flow algorithm. The similarity of two fingerprints is evaluated using the number of common matching pairs and the maximal posteriori probability. In order to deal with part-matching problems, we use the smallest convex hull which contains all the matched minutiae. Experiments evaluated on FVC 2004 show both effectiveness and efficiency of our methods.

This paper proposes a hybrid method for face recognition using local features and statistical feature extraction methods. First, a dense set of local feature points are extracted in order to represent a facial image. Each local feature point is described by the keypoint descriptor defined by SIFT feature. Then, the statistical feature extraction methods, PCA and LDA, are applied to the set of local feature descriptors in order to find low dimensional features. With the obtained low dimensional feature vectors, we can conduct face recognition task efficiently using a simple classifier. Through computational experiments on benchmark data sets, we show that the proposed method is superior to the conventional PCA and LDA in the classification performance. In addition, we also show that the proposed method can achieve remarkable improvement in the processing time compared to the conventional keypoint matching methods proposed for local features.KeywordsFace RecognitionLocal FeatureSift FeatureGray Level IntensityConventional Statistical MethodThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

The interpretation of prostate MRI suffers from low agreement across radiologists due to the subtle differences between cancer and normal tissue. Image registration addresses this issue by accurately mapping the ground-truth cancer labels from surgical histopathology images onto MRI. Cancer labels achieved by image registration can be used to improve radiologists’ interpretation of MRI by training deep learning models for early detection of prostate cancer. A major limitation of current automated registration approaches is that they require manual prostate segmentations, which is a time-consuming task, prone to errors. This paper presents a weakly supervised approach for affine and deformable registration of MRI and histopathology images without requiring prostate segmentations. We used manual prostate segmentations and mono-modal synthetic image pairs to train our registration networks to align prostate boundaries and local prostate features. Although prostate segmentations were used during the training of the network, such segmentations were not needed when registering unseen images at inference time. We trained and validated our registration network with 135 and 10 patients from an internal cohort, respectively. We tested the performance of our method using 16 patients from the internal cohort and 22 patients from an external cohort. The results show that our weakly supervised method has achieved significantly higher registration accuracy than a state-of-the-art method run without prostate segmentations. Our deep learning framework will ease the registration of MRI and histopathology images by obviating the need for prostate segmentations.

In this study, a visual similarity metric based on precision---recall graphs is presented as an alternative to the widely used Hausdorff distance (HD). Such metric, called maximum cardinality similarity metric, is computed between a reference shape and a test template, each one represented by a set of edge points. We address this problem using a bipartite graph representation of the relationship between the sets. The matching problem is solved using the Hopcroft---Karp algorithm, taking advantage of its low computational complexity. We present a comparison between our results and those obtained from applying the partial Hausdorff distance (PHD) to the same test sets. Similar results were found using both approaches for standard template-matching applications. Nevertheless, the proposed methodology is more accurate at determining the completeness of partial shapes under noise conditions. Furthermore, the processing time required by our methodology is lower than that required to compute the PHD, for a large set of points.

In the approximate Euclidean min-weighted perfect matching problem, a set of points in the plane and a real number are given. Usually, a solution of this problem is a partition of points of into pairs such that the sum of the distances between the paired points is at most times the optimal solution. In this paper, the authors give a randomized algorithm which follows a Monte-Carlo method. This algorithm is a randomized fully polynomial-time approximation scheme for the given problem. Fortunately, the suggested algorithm is a one tackled the matching problem in both Euclidean nonbipartite and bipartite cases. The presented algorithm outlines as follows: With repeating times, we choose a point from to build the suitable pair satisfying the suggested condition on the distance. If this condition is achieved, then remove the points of the constructed pair from and put this pair in (the output set of the solution). Then, choose a point and the nearest point of it from the remaining points in to construct a pair and put it in . Remove the two points of the constructed pair from and repeat this process until becomes an empty set. Obviously, this method is very simple. Furthermore, our algorithm can be applied without any modification on complete weighted graphs and complete weighted bipartite graphs , where and m is an even.

In this work, we propose a parameterised quantum circuit learning approach to point set matching problem. In contrast to previous annealing-based methods, we propose a quantum circuit-based frame-work whose parameters are optimised via descending the gradients w.r.t a kernel-based loss function. We formulate the shape matching problem into a distribution learning task; that is, to learn the distribution of the optimal transformation parameters. We show that this framework is able to find multiple optimal solutions for symmetric shapes and is more accurate, scalable and robust than the previous annealing-based method.

This paper proposes a new affine registration algorithm for 2D point matching. It is a two-step iterative registration algorithm by soft weight assignment based on bidirectional distance. At each iteration, the affine transformation is updated by two optimization steps, in which the model data and the test data are matched from each other respectively. By the optimization of registration at separate steps during each iteration, the proposed algorithm can provide a good estimate of the accurate affine transformation in condition of poor initialization and lack of geometric assumptions on point sets. Some experiments about comparison with the current state-of-the art approaches, demonstrate the robustness and accuracy of our method.