Diffusion generated methods for denoising target-valued images
From MaRDI portal
denoisingDT-MRIimage analysisline fieldmanifold-valued dataMerriman-Bence-Osher (MBO) diffusion generated motion
Heat equation (35K05) PDEs in connection with classical thermodynamics and heat transfer (35Q79) Integro-partial differential equations (35R09) PDEs with randomness, stochastic partial differential equations (35R60) Manifolds and measure-geometric topics (49Q99) Numerical aspects of computer graphics, image analysis, and computational geometry (65D18) Numerical optimization and variational techniques (65K10) Stability and convergence of numerical methods for initial value and initial-boundary value problems involving PDEs (65M12) Numerical methods for inverse problems for initial value and initial-boundary value problems involving PDEs (65M32)
Abstract: We consider the inverse problem of denoising an image where each point (pixel) is an element of a target set, which we refer to as a target-valued image. The target sets considered are either (i) a closed convex set of Euclidean space or (ii) a closed subset of the sphere such that the closest point mapping is defined almost everywhere. The energy for the denoising problem consists of an -fidelity term which is regularized by the Dirichlet energy. A relaxation of this energy, based on the heat kernel, is introduced and the associated minimization problem is proven to be well-posed. We introduce a diffusion generated method which can be used to efficiently find minimizers of this energy. We prove results for the stability and convergence of the method for both types of target sets. The method is demonstrated on a variety of synthetic and test problems, with associated target sets given by the semi-positive definite matrices, the cube, spheres, the orthogonal matrices, and the real projective line.
Recommendations
- scientific article; zbMATH DE number 5077546
- Image denoising using generalized anisotropic diffusion
- Image denoising and segmentation via nonlinear diffusion
- Image denoising via multiscale nonlinear diffusion models
- scientific article; zbMATH DE number 1918727
- Image denoising via nonlinear hybrid diffusion
- A new diffusion-based variational model for image denoising and segmentation
- Efficient diffusion coefficient for image denoising
Cites work
- A diffusion generated method for computing Dirichlet partitions
- A diffusion generated method for orthogonal matrix-valued fields
- A Graph Framework for Manifold-Valued Data
- A minimal surface criterion for graph partitioning
- A non-local anisotropic model for phase transitions: asymptotic behaviour of rescaled energies
- A nonlocal denoising algorithm for manifold-valued images using second order statistics
- A second order nonsmooth variational model for restoring manifold-valued images
- A simple embedding method for solving partial differential equations on surfaces
- An Approach to Quad Meshing Based on Harmonic Cross-Valued Maps and the Ginzburg--Landau Theory
- An efficient iterative thresholding method for image segmentation
- An efficient threshold dynamics method for wetting on rough surfaces
- An improved threshold dynamics method for wetting dynamics
- An optimal partition problem for eigenvalues
- Analysis of diffusion generated motion for mean curvature flow in codimension two: a gradient-flow approach
- Consistency of Dirichlet partitions
- Convergence of the thresholding scheme for multi-phase mean-curvature flow
- Convergence of thresholding schemes incorporating bulk effects
- Convex optimization algorithms
- Diffuse interface models on graphs for classification of high dimensional data
- Diffusion generated motion using signed distance functions
- Diffusion-generated motion by mean curvature for filaments
- Ginzburg-Landau vortices
- Graph clustering, variational image segmentation methods and Hough transform scale detection for object measurement in images
- scientific article; zbMATH DE number 1376935 (Why is no real title available?)
- Interface dynamics for an Allen-Cahn-type equation governing a matrix-valued field
- Landau-De Gennes theory of nematic liquid crystals: the Oseen-Frank limit and beyond
- Mean curvature, threshold dynamics, and phase field theory on finite graphs
- Minimal Dirichlet energy partitions for graphs
- Minimization of Region-Scalable Fitting Energy for Image Segmentation
- Modelling and control of robot manipulators.
- Nonlinear total variation based noise removal algorithms
- Numerical methods for the Landau--Lifshitz equation
- Reaction-Diffusion Processes and Evolution to Harmonic Maps
- Scattered manifold-valued data approximation
- Short-time heat flow and functions of bounded variation in \(\mathbb R^N\)
- Simple computation of reaction-diffusion processes on point clouds
- Threshold dynamics for high order geometric motions
- Threshold dynamics for networks with arbitrary surface tensions
- Threshold dynamics for the piecewise constant Mumford-Shah functional
- Total variation regularization for manifold-valued data
- Variational Analysis in Sobolev andBVSpaces
Cited in
(18)- An efficient iterative method for reconstructing surface from point clouds
- Diffusion tensor regularization with metric double integrals
- Mapping-based image diffusion
- Fast operator splitting methods for obstacle problems
- An Approach to Quad Meshing Based on Harmonic Cross-Valued Maps and the Ginzburg--Landau Theory
- An Efficient Unconditionally Stable Method for Dirichlet Partitions in Arbitrary Domains
- A diffusion generated method for orthogonal matrix-valued fields
- Interface dynamics for an Allen-Cahn-type equation governing a matrix-valued field
- A Characteristic Function-Based Algorithm for Geodesic Active Contours
- An Iterative Thresholding Method for Topology Optimization for the Navier–Stokes Flow
- A prediction-correction based iterative convolution-thresholding method for topology optimization of heat transfer problems
- Maximum bound principle for matrix-valued Allen-Cahn equation and integrating factor Runge-Kutta method
- Denoising of sphere- and SO(3)-valued data by relaxed Tikhonov regularization
- Efficient and stable diffusion generated methods for ground state computation in Bose-Einstein condensates
- On the maximum bound principle and energy dissipation of exponential time differencing methods for the matrix-valued Allen-Cahn equation
- Unconditional energy dissipation of Strang splitting for the matrix-valued Allen-Cahn equation
- An efficient threshold dynamics method for topology optimization for fluids
- An efficient iterative convolution-thresholding method for image inpainting
This page was built for publication: Diffusion generated methods for denoising target-valued images
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2176514)
