Showing papers in "Journal of Mathematical Imaging and Vision in 2018"

Video Denoising via Empirical Bayesian Estimation of Space-Time Patches

Abstract: In this paper we present a new patch-based empirical Bayesian video denoising algorithm. The method builds a Bayesian model for each group of similar space-time patches. These patches are not motion-compensated, and therefore avoid the risk of inaccuracies caused by motion estimation errors. The high dimensionality of spatiotemporal patches together with a limited number of available samples poses challenges when estimating the statistics needed for an empirical Bayesian method. We therefore assume that groups of similar patches have a low intrinsic dimensionality, leading to a spiked covariance model. Based on theoretical results about the estimation of spiked covariance matrices, we propose estimators of the eigenvalues of the a priori covariance in high-dimensional spaces as simple corrections of the eigenvalues of the sample covariance matrix. We demonstrate empirically that these estimators lead to better empirical Wiener filters. A comparison on classic benchmark videos demonstrates improved visual quality and an increased PSNR with respect to state-of-the-art video denoising methods.

Normal Integration: A Survey

Abstract: The need for efficient normal integration methods is driven by several computer vision tasks such as shape-from-shading, photometric stereo, deflectometry. In the first part of this survey, we select the most important properties that one may expect from a normal integration method, based on a thorough study of two pioneering works by Horn and rooks (Comput Vis Graph Image Process 33(2): 174–208, 1986) and Frankot and Chellappa (IEEE Trans Pattern Anal Mach Intell 10(4): 439-451, 1988). Apart from accuracy, an integration method should at least be fast and robust to a noisy normal field. In addition, it should be able to handle several types of boundary condition, including the case of a free boundary and a reconstruction domain of any shape, i.e., which is not necessarily rectangular. It is also much appreciated that a minimum number of parameters have to be tuned, or even no parameter at all. Finally, it should preserve the depth discontinuities. In the second part of this survey, we review most of the existing methods in view of this analysis and conclude that none of them satisfies all of the required properties. This work is complemented by a companion paper entitled Variational Methods for Normal Integration, in which we focus on the problem of normal integration in the presence of depth discontinuities, a problem which occurs as soon as there are occlusions.

LED-based Photometric Stereo: Modeling, Calibration and Numerical Solution

Abstract: We conduct a thorough study of photomet-ric stereo under nearby point light source illumination, from modeling to numerical solution, through calibration. In the classical formulation of photometric stereo, the luminous uxes are assumed to be directional, which is very dicult to achieve in practice. Rather, we use light-emitting diodes (LEDs) to illuminate the scene to be reconstructed. Such point light sources are very convenient to use, yet they yield a more complex photo-metric stereo model which is arduous to solve. We rst derive in a physically sound manner this model, and show how to calibrate its parameters. Then, we discuss two state-of-the-art numerical solutions. The rst one alternatingly estimates the albedo and the normals, and then integrates the normals into a depth map. It is shown empirically to be independent from the ini-tialization, but convergence of this sequential approach is not established. The second one directly recovers the depth, by formulating photometric stereo as a system of nonlinear partial dierential equations (PDEs), which are linearized using image ratios. Although the sequential approach is avoided, initialization matters a lot and convergence is not established either. Therefore, we introduce a provably convergent alternating reweighted least-squares scheme for solving the original system of nonlinear PDEs. Finally, we extend this study to the case of RGB images.

Hierarchical Segmentations with Graphs: Quasi-flat Zones, Minimum Spanning Trees, and Saliency Maps

Abstract: Hierarchies of partitions are generally represented by dendrograms (direct representation). They can also be represented by saliency maps or minimum spanning trees. In this article, we precisely study the links between these three representations. In particular, we provide a new bijection between saliency maps and hierarchies based on quasi-flat zones as often used in image processing and we characterize saliency maps and minimum spanning trees as solutions to constrained minimization problems where the constraint is quasi-flat zones preservation. In practice, these results make up a toolkit for designing new hierarchical methods where one can choose the most convenient representation. They also invite us to process non-image data with morphological hierarchies. More precisely, we show the practical interest of the proposed framework for: i) hierarchical watershed image segmentations, ii) combinations of dierent hierarchical segmentations, iii) hierarchicalizations of some non-hierarchical image segmentation methods based on regional dissimilarities, and iv) hierarchical analysis of geographical data.

Path-Value Functions for Which Dijkstra’s Algorithm Returns Optimal Mapping

Abstract: Dijkstra’s algorithm (DA) is one of the most useful and efficient graph-search algorithms, which can be modified to solve many different problems. It is usually presented as a tool for finding a mapping which, for every vertex v, returns a shortest-length path to v from a fixed single source vertex. However, it is well known that DA returns also a correct optimal mapping when multiple sources are considered and for path-value functions more general than the standard path-length. The use of DA in such general setting can reduce many image processing operations to the computation of an optimum-path forest with path-cost function defined in terms of local image attributes. In this paper, we describe the general properties of a path-value function defined on an arbitrary finite graph which, provably, ensure that Dijkstra’s algorithm indeed returns an optimal mapping. We also provide the examples showing that the properties presented in a 2004 TPAMI paper on the image foresting transform, which were supposed to imply proper behavior of DA, are actually insufficient. Finally, we describe the properties of the path-value function of a graph that are provably necessary for the algorithm to return an optimal mapping.

Optimal Paths for Variants of the 2D and 3D Reeds–Shepp Car with Applications in Image Analysis

Abstract: We present a PDE-based approach for finding optimal paths for the Reeds–Shepp car. In our model we minimize a (data-driven) functional involving both curvature and length penalization, with several generalizations. Our approach encompasses the two- and three-dimensional variants of this model, state-dependent costs, and moreover, the possibility of removing the reverse gear of the vehicle. We prove both global and local controllability results of the models. Via eikonal equations on the manifold $$\mathbb {R}^d \times {\mathbb {S}}^{d-1}$$ we compute distance maps w.r.t. highly anisotropic Finsler metrics, which approximate the singular (quasi)-distances underlying the model. This is achieved using a fast-marching (FM) method, building on Mirebeau (Numer Math 126(3):515–557, 2013; SIAM J Numer Anal 52(4):1573–1599, 2014). The FM method is based on specific discretization stencils which are adapted to the preferred directions of the Finsler metric and obey a generalized acuteness property. The shortest paths can be found with a gradient descent method on the distance map, which we formalize in a theorem. We justify the use of our approximating metrics by proving convergence results. Our curve optimization model in $$\mathbb {R}^{d} \times \mathbb {S}^{d-1}$$ with data-driven cost allows to extract complex tubular structures from medical images, e.g., crossings, and incomplete data due to occlusions or low contrast. Our work extends the results of Sanguinetti et al. (Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications LNCS 9423, 2015) on numerical sub-Riemannian eikonal equations and the Reeds–Shepp car to 3D, with comparisons to exact solutions by Duits et al. (J Dyn Control Syst 22(4):771–805, 2016). Numerical experiments show the high potential of our method in two applications: vessel tracking in retinal images for the case $$d=2$$ and brain connectivity measures from diffusion-weighted MRI data for the case $$d=3$$ , extending the work of Bekkers et al. (SIAM J Imaging Sci 8(4):2740–2770, 2015). We demonstrate how the new model without reverse gear better handles bifurcations.

Fast-Marching Methods for Curvature Penalized Shortest Paths

Abstract: We introduce numerical schemes for computing distances and shortest paths with respect to several planar paths models, featuring curvature penalization and data-driven velocity: the Dubins car, the Euler/Mumford elastica, and a two variants of the Reeds–Shepp car. For that purpose, we design monotone and causal discretizations of the associated Hamilton–Jacobi–Bellman PDEs, posed on the three-dimensional domain $${\mathbb R}^2 \times {\mathbb S}^1$$ . Our discretizations involve sparse, adaptive and anisotropic stencils on a cartesian grid, built using techniques from lattice geometry. A convergence proof is provided, in the setting of discontinuous viscosity solutions. The discretized problems are solvable in a single pass using a variant of the fast-marching algorithm. Numerical experiments illustrate the applications of our schemes in motion planning and image segmentation.

Rendering Deformed Speckle Images with a Boolean Model

Abstract: Rendering speckle images affected by a given deformation field is of primary importance to assess the metrological performance of displacement measurement methods used in experimental mechanics and based on digital image correlation (DIC). This article describes how to render deformed speckle images with a classic model of stochastic geometry, the Boolean model. The advantage of the proposed approach is that it does not depend on any interpolation scheme likely to bias the assessment process, and that it allows the user to render speckle images deformed with any deformation field given by an analytic formula. The proposed algorithm mimics the imaging chain of a digital camera, and its parameters are carefully discussed. A MATLAB software implementation and synthetic ground-truth datasets for assessing DIC software programs are publicly available.

Fast Recursive Computation of Krawtchouk Polynomials

Abstract: Krawtchouk polynomials (KPs) and their moments are used widely in the field of signal processing for their superior discriminatory properties. This study proposes a new fast recursive algorithm to compute Krawtchouk polynomial coefficients (KPCs). This algorithm is based on the symmetry property of KPCs along the primary and secondary diagonals of the polynomial array. The $$n-x$$ plane of the KP array is partitioned into four triangles, which are symmetrical across the primary and secondary diagonals. The proposed algorithm computes the KPCs for only one triangle (partition), while the coefficients of the other three triangles (partitions) can be computed using the derived symmetry properties of the KP. Therefore, only N / 4 recursion times are required. The proposed algorithm can also be used to compute polynomial coefficients for different values of the parameter p in interval (0, 1). The performance of the proposed algorithm is compared with that in previous literature in terms of image reconstruction error, polynomial size, and computation cost. Moreover, the proposed algorithm is applied in a face recognition system to determine the impact of parameter p on feature extraction ability. Simulation results show that the proposed algorithm has a remarkable advantage over other existing algorithms for a wide range of parameters p and polynomial size N, especially in reducing the computation time and the number of operations utilized.

A Tutorial on Well-Composedness

Abstract: Due to digitization, usual discrete signals generally present topological paradoxes, such as the connectivity paradoxes of Rosenfeld. To get rid of those paradoxes, and to restore some topological properties to the objects contained in the image, like manifoldness, Latecki proposed a new class of images, called well-composed images, with no topological issues. Furthermore, well-composed images have some other interesting properties: for example, the Euler number is locally computable, boundaries of objects separate background from foreground, the tree of shapes is well defined. Last, but not the least, some recent works in mathematical morphology have shown that very nice practical results can be obtained thanks to well-composed images. Believing in its prime importance in digital topology, we then propose this state of the art of well-composedness, summarizing its different flavors, the different methods existing to produce well-composed signals, and the various topics that are related to well-composedness.

Modified Rodrigues Parameters: An Efficient Representation of Orientation in 3D Vision and Graphics

Abstract: Modified Rodrigues parameters (MRPs) are triplets in $${\mathbb {R}}^3$$R3 bijectively and rationally mapped to quaternions through stereographic projection. We present here a compelling case for MRPs as a minimal degree-of-freedom parameterization of orientation through novel solutions to prominent problems in the fields of 3D vision and computer graphics. In our primary contribution, we show that the derivatives of a unit quaternion in terms of its MRPs are simple polynomial expressions of its scalar and vector part. Furthermore, we show that updates to unit quaternions from perturbations in parameter space can be computed without explicitly invoking the parameters in the computations. Based on the former, we introduce a novel approach for designing orientation splines by configuring their back-projections in 3D space. Finally, in the general topic of nonlinear optimization for geometric vision, we run performance analyses and provide comparisons on the convergence behavior of MRP parameterizations on the tasks of absolute orientation, exterior orientation and large-scale bundle adjustment of public datasets.

New Set of Quaternion Moments for Color Images Representation and Recognition

Abstract: In this paper, a new set of quaternion radial-substituted Chebyshev moments (QRSCMs) is proposed for color image representation and recognition. These new moments are circular moments defined over a unit disk by using a new set of orthogonal basis functions called radial-substituted Chebyshev functions. A new hybrid method is proposed for highly accurate computation of QRSCMs in polar coordinates. In this method, the angular kernel is exactly computed by analytical integration of Fourier function over circular pixels. The radial kernel is computed using a recurrence relation which completely eliminates the coefficient matrix associated with the radial-substituted Chebyshev functions. Rotation, scaling, and translation (RST) invariances for QRSCMs are proved. Numerical experiments were conducted where the results of these experiments show better performance of QRSCMs over existing quaternion moments in terms of image reconstruction capabilities, RST invariances, robust to different noises, and CPU elapsed times.

A Neuromathematical Model for Geometrical Optical Illusions

Abstract: Geometrical optical illusions have been object of many studies due to the possibility they offer to understand the behavior of low-level visual processing. They consist in situations in which the perceived geometrical properties of an object differ from those of the object in the visual stimulus. Starting from the geometrical model introduced by Citti and Sarti (J Math Imaging Vis 24(3):307---326, 2006), we provide a mathematical model and a computational algorithm which allows to interpret these phenomena and to qualitatively reproduce the perceived misperception.

Variational Methods for Normal Integration

Abstract: The need for an efficient method of integration of a dense normal field is inspired by several computer vision tasks, such as shape-from-shading, photometric stereo, deflectometry. Inspired by edge-preserving methods from image processing, we study in this paper several variational approaches for normal integration, with a focus on non-rectangular domains, free boundary and depth discontinuities. We first introduce a new discretization for quadratic integration, which is designed to ensure both fast recovery and the ability to handle non-rectangular domains with a free boundary. Yet, with this solver, discontinuous surfaces can be handled only if the scene is first segmented into pieces without discontinuity. Hence, we then discuss several discontinuity-preserving strategies. Those inspired, respectively, by the Mumford–Shah segmentation method and by anisotropic diffusion, are shown to be the most effective for recovering discontinuities.

Priors with Coupled First and Second Order Differences for Manifold-Valued Image Processing

Abstract: We generalize discrete variational models involving the infimal convolution (IC) of first and second order differences and the total generalized variation (TGV) to manifold-valued images. We propose both extrinsic and intrinsic approaches. The extrinsic models are based on embedding the manifold into an Euclidean space of higher dimension with manifold constraints. An alternating direction methods of multipliers can be employed for finding the minimizers. However, the components within the extrinsic IC or TGV decompositions live in the embedding space which makes their interpretation difficult. Therefore, we investigate two intrinsic approaches: for Lie groups, we employ the group action within the models; for more general manifolds, our IC model is based on recently developed absolute second order differences on manifolds, while our TGV approach uses an approximation of the parallel transport by the pole ladder. For computing the minimizers of the intrinsic models, we apply gradient descent algorithms. Numerical examples demonstrate that our approaches work well for certain manifolds.

Image Segmentation with Depth Information via Simplified Variational Level Set Formulation

Abstract: Image segmentation with depth information can be modeled as a minimization problem with Nitzberg–Mumford–Shiota functional, which can be transformed into a tractable variational level set formulation. However, such formulation leads to a series of complicated high-order nonlinear partial differential equations which are difficult to solve efficiently. In this paper, we first propose an equivalently reduced variational level set formulation without using curvatures by taking level set functions as signed distance functions. Then, an alternating direction method of multipliers (ADMM) based on this simplified variational level set formulation is designed by introducing some auxiliary variables, Lagrange multipliers via using alternating optimization strategy. With the proposed ADMM method, the minimization problem for this simplified variational level set formulation is transformed into a series of sub-problems, which can be solved easily via using the Gauss–Seidel iterations, fast Fourier transform and soft thresholding formulas. The level set functions are treated as signed distance functions during computation process via implementing a simple algebraic projection method, which avoids the traditional re-initialization process for conventional variational level set methods. Extensive experiments have been conducted on both synthetic and real images, which validate the proposed approach, and show advantages of the proposed ADMM projection over algorithms based on traditional gradient descent method in terms of computational efficiency.

Rate-Invariant Analysis of Covariance Trajectories

Abstract: Statistical analysis of dynamic systems, such as videos and dynamic functional connectivity, is often translated into a problem of analyzing trajectories of relevant features, particularly covariance matrices. As an example, in video-based action recognition, a natural mathematical representation of activity videos is as parameterized trajectories on the set of symmetric, positive-definite matrices (SPDMs). The execution rates of actions, implying arbitrary parameterizations of trajectories, complicate their analysis. To handle this challenge, we represent covariance trajectories using transported square-root vector fields, constructed by parallel translating scaled-velocity vectors of trajectories to their starting points. The space of such representations forms a vector bundle on the SPDM manifold. Using a natural Riemannian metric on this vector bundle, we approximate geodesic paths and geodesic distances between trajectories in the space of this vector bundle. This metric is invariant to the action of the re-parameterization group, and leads to a rate-invariant analysis of trajectories. In the process, we remove the parameterization variability and temporally register trajectories. We demonstrate this framework in multiple contexts, using both generative statistical models and discriminative data analysis. The latter is illustrated using several applications involving video-based action recognition and dynamic functional connectivity analysis.

A Novel Diffeomorphic Model for Image Registration and Its Algorithm

Abstract: In this work, we investigate image registration by mapping one image to another in a variational framework and focus on both model robustness and solver efficiency. We first propose a new variational model with a special regularizer, based on the quasi-conformal theory, which can guarantee that the registration map is diffeomorphic. It is well known that when the deformation is large, many variational models including the popular diffusion model cannot ensure diffeomorphism. One common observation is that the fidelity error appears small while the obtained transform is incorrect by way of mesh folding. However, direct reformulation from the Beltrami framework does not lead to effective models; our new regularizer is constructed based on this framework and added to the diffusion model to get a new model, which can achieve diffeomorphism. However, the idea is applicable to a wide class of models. We then propose an iterative method to solve the resulting nonlinear optimization problem and prove the convergence of the method. Numerical experiments can demonstrate that the new model can not only get a diffeomorphic registration even when the deformation is large, but also possess the accuracy in comparing with the currently best models.

Design and Processing of Invertible Orientation Scores of 3D Images.

Abstract: The enhancement and detection of elongated structures in noisy image data are relevant for many biomedical imaging applications. To handle complex crossing structures in 2D images, 2D orientation scores $$U: {\mathbb {R}} ^ 2\times S ^ 1 \rightarrow {\mathbb {C}}$$ were introduced, which already showed their use in a variety of applications. Here we extend this work to 3D orientation scores $$U: {\mathbb {R}} ^ 3 \times S ^ 2\rightarrow {\mathbb {C}}$$ . First, we construct the orientation score from a given dataset, which is achieved by an invertible coherent state type of transform. For this transformation we introduce 3D versions of the 2D cake wavelets, which are complex wavelets that can simultaneously detect oriented structures and oriented edges. Here we introduce two types of cake wavelets: the first uses a discrete Fourier transform, and the second is designed in the 3D generalized Zernike basis, allowing us to calculate analytical expressions for the spatial filters. Second, we propose a nonlinear diffusion flow on the 3D roto-translation group: crossing-preserving coherence-enhancing diffusion via orientation scores (CEDOS). Finally, we show two applications of the orientation score transformation. In the first application we apply our CEDOS algorithm to real medical image data. In the second one we develop a new tubularity measure using 3D orientation scores and apply the tubularity measure to both artificial and real medical data.

Image Denoising Using Generalized Anisotropic Diffusion

Abstract: Motivated by some recent works in fractional-order anisotropic diffusions, we introduce a class of generalized anisotropic diffusion equations for image denoising We first define a new type of derivative (called G-derivative), which contains the fractional derivative as a special case, using the Fourier transform, then the generalized anisotropic diffusion equations are Euler–Lagrange equations of a cost functional which is an increasing function of the absolute value of the G-derivative of the image intensity function All the G-derivative operators constitute a ring, and the semigroup property of the G-derivative consists with the semigroup property of the fractional derivative, so the resulting generalized anisotropic diffusions can be seen as generalizations of the fractional-order anisotropic diffusions We also discuss the generalized Sobolev space described by the G-derivative and some variants of generalized anisotropic diffusions The discretization of the G-derivative is computed in the frequency domain, and the stability analysis of the difference scheme is given We list some generalized anisotropic diffusions and apply them to image denoising Numerical results show that new models have great potentials in image denoising

Iterative Regularization via Dual Diagonal Descent

Abstract: In the context of linear inverse problems, we propose and study a general iterative regularization method allowing to consider large classes of data-fit terms and regularizers. The algorithm we propose is based on a primal-dual diagonal descent method. Our analysis establishes convergence as well as stability results. Theoretical findings are complemented with numerical experiments showing state-of-the-art performances.

Highly Corrupted Image Inpainting Through Hypoelliptic Diffusion

Abstract: We present a new image inpainting algorithm, the Averaging and Hypoelliptic Evolution (AHE) algorithm, inspired by the one presented in [1] and based upon a (semi-discrete) variation of the Citti--Petitot--Sarti model of the primary visual cortex V1. In particular, we focus on reconstructing highly corrupted images (i.e. where more than the 80% of the image is missing). [1] U. Boscain, R. A. Chertovskih, J. P. Gauthier, and A. O. Remizov, Hypoelliptic diffusion and human vision: a semidiscrete new twist, SIAM J. Imaging Sci., vol. 7, no. 2, pp. 669--695, 2014.

Topological Fidelity and Image Thresholding: A Persistent Homology Approach

Abstract: We develop a method based on persistent homology to analyze topological structure in noisy digital images. The method returns threshold(s) for image segmentation to represent inherent topological structure as well as estimates of topological quantities in the form of Betti numbers. Two motivating data sets are scans of binary alloys and firn, the intermediate stage between snow and ice.

Spatially Constrained Student’s t-Distribution Based Mixture Model for Robust Image Segmentation

Abstract: The finite Gaussian mixture model is one of the most popular frameworks to model classes for probabilistic model-based image segmentation. However, the tails of the Gaussian distribution are often shorter than that required to model an image class. Also, the estimates of the class parameters in this model are affected by the pixels that are atypical of the components of the fitted Gaussian mixture model. In this regard, the paper presents a novel way to model the image as a mixture of finite number of Student’s t-distributions for image segmentation problem. The Student’s t-distribution provides a longer tailed alternative to the Gaussian distribution and gives reduced weight to the outlier observations during the parameter estimation step in finite mixture model. Incorporating the merits of Student’s t-distribution into the hidden Markov random field framework, a novel image segmentation algorithm is proposed for robust and automatic image segmentation, and the performance is demonstrated on a set of HEp-2 cell and natural images. Integrating the bias field correction step within the proposed framework, a novel simultaneous segmentation and bias field correction algorithm has also been proposed for segmentation of magnetic resonance (MR) images. The efficacy of the proposed approach, along with a comparison with related algorithms, is demonstrated on a set of real and simulated brain MR images both qualitatively and quantitatively.

Nonlocal Myriad Filters for Cauchy Noise Removal

Abstract: The contribution of this paper is twofold. First, we introduce a generalized myriad filter, which is a method to compute the joint maximum likelihood estimator of the location and the scale parameter of the Cauchy distribution. Estimating only the location parameter is known as myriad filter. We propose an efficient algorithm to compute the generalized myriad filter and prove its convergence. Special cases of this algorithm result in the classical myriad filtering and an algorithm for estimating only the scale parameter. Based on an asymptotic analysis, we develop a second, even faster generalized myriad filtering technique. Second, we use our new approaches within a nonlocal, fully unsupervised method to denoise images corrupted by Cauchy noise. Special attention is paid to the determination of similar patches in noisy images. Numerical examples demonstrate the excellent performance of our algorithms which have moreover the advantage to be robust with respect to the parameter choice.

Whiteness Constraints in a Unified Variational Framework for Image Restoration

Abstract: We propose a robust variational model for the restoration of images corrupted by blur and the general class of additive white noises. The key idea behind our proposal relies on a novel hard constraint imposed on the residual of the restoration, namely we characterize a residual whiteness set to which the restored image must belong. As the feasible set is unbounded, solution existence results for the proposed variational model are given. Moreover, based on theoretical derivations as well as on Monte Carlo simulations, we provide well-founded guidelines for setting the whiteness constraint limits. The solution of the non-trivial optimization problem, due to the non-smooth non-convex proposed model, is efficiently obtained by an alternating directions method of multipliers, which in particular reduces the solution to a sequence of convex optimization subproblems. Numerical results show the potentiality of the proposed model for restoring blurred images corrupted by several kinds of additive white noises.

Topology-Preserving Image Segmentation by Beltrami Representation of Shapes

Abstract: A new approach using the Beltrami representation of a shape for topology-preserving image segmentation is proposed in this paper. Using the proposed model, the target object can be segmented from the input image by a region of user-prescribed topology. Given a target image I, a template image J is constructed and then deformed with respect to the Beltrami representation. The deformation on J is designed such that the topology of the segmented region is preserved as which the object is interior in J. The topology-preserving property of the deformation is guaranteed by imposing only one constraint on the Beltrami representation, which is easy to be handled. Introducing the Beltrami representation also allows large deformations on the topological prior J, so that it can be a very simple image, such as an image of disks, torus, disjoint disks. Hence, prior shape information of I is unnecessary for the proposed model. Additionally, the proposed model can be easily incorporated with selective segmentation, in which landmark constraints can be imposed interactively to meet any practical need (e.g., medical imaging). High accuracy and stability of the proposed model to deal with different segmentation tasks are validated by numerical experiments on both artificial and real images.

Fast Asymmetric Fronts Propagation for Image Segmentation

Abstract: In this paper, we introduce a generalized asymmetric fronts propagation model based on the geodesic distance maps and the Eikonal partial differential equations. One of the key ingredients for the computation of the geodesic distance map is the geodesic metric, which can govern the action of the geodesic distance level set propagation. We consider a Finsler metric with the Randers form, through which the asymmetry and anisotropy enhancements can be taken into account to prevent the fronts leaking problem during the fronts propagation. These enhancements can be derived from the image edge-dependent vector field such as the gradient vector flow. The numerical implementations are carried out by the Finsler variant of the fast marching method, leading to very efficient interactive segmentation schemes. We apply the proposed Finsler fronts propagation model to image segmentation applications. Specifically, the foreground and background segmentation is implemented by the Voronoi index map. In addition, for the application of tubularity segmentation, we exploit the level set lines of the geodesic distance map associated with the proposed Finsler metric providing that a thresholding value is given.

Nilpotent Approximations of Sub-Riemannian Distances for Fast Perceptual Grouping of Blood Vessels in 2D and 3D

Abstract: We propose an efficient approach for the grouping of local orientations (points on vessels) via nilpotent approximations of sub-Riemannian distances in the 2D and 3D roto-translation groups SE(2) and SE(3). In our distance approximations we consider homogeneous norms on nilpotent groups that locally approximate SE(n), and which are obtained via the exponential and logarithmic map on SE(n). In a qualitative validation we show that the norms provide accurate approximations of the true sub-Riemannian distances, and we discuss their relations to the fundamental solution of the sub-Laplacian on SE(n). The quantitative experiments further confirm the accuracy of the approximations. Quantitative results are obtained by evaluating perceptual grouping performance of retinal blood vessels in 2D images and curves in challenging 3D synthetic volumes. The results show that (1) sub-Riemannian geometry is essential in achieving top performance and (2) grouping via the fast analytic approximations performs almost equally, or better, than data-adaptive fast marching approaches on $$\mathbb {R}^n$$ and SE(n).

A Geometric Model of Multi-scale Orientation Preference Maps via Gabor Functions

Abstract: In this paper we present a new model for the generation of orientation preference maps in the primary visual cortex (V1), considering both orientation and scale features. First we undertake to model the functional architecture of V1 by interpreting it as a principal fiber bundle over the 2-dimensional retinal plane by introducing intrinsic variables orientation and scale. The intrinsic variables constitute a fiber on each point of the retinal plane and the set of receptive profiles of simple cells is located on the fiber. Each receptive profile on the fiber is mathematically interpreted as a rotated Gabor function derived from an uncertainty principle. The visual stimulus is lifted in a 4-dimensional space, characterized by coordinate variables, position, orientation and scale, through a linear filtering of the stimulus with Gabor functions. Orientation preference maps are then obtained by mapping the orientation value found from the lifting of a noise stimulus onto the 2-dimensional retinal plane. This corresponds to a Bargmann transform in the reducible representation of the $$\text {SE}(2)=\mathbb {R}^2\times S^1$$ group. A comparison will be provided with a previous model based on the Bargmann transform in the irreducible representation of the $$\text {SE}(2)$$ group, outlining that the new model is more physiologically motivated. Then, we present simulation results related to the construction of the orientation preference map by using Gabor filters with different scales and compare those results to the relevant neurophysiological findings in the literature.