First Order Algorithms in Variational Image Processing

DOI10.1007/978-3-319-41589-5_10zbMATH Open1372.65053arXiv1412.4237OpenAlexW2125676375MaRDI QIDQ5350482FDOQ5350482

Gabriele Steidl, Martin Burger, Alex Sawatzky

Publication date: 1 September 2017

Published in: Splitting Methods in Communication, Imaging, Science, and Engineering (Search for Journal in Brave)

Abstract: Variational methods in imaging are nowadays developing towards a quite universal and flexible tool, allowing for highly successful approaches on tasks like denoising, deblurring, inpainting, segmentation, super-resolution, disparity, and optical flow estimation. The overall structure of such approaches is of the form $calD(Ku)+alphacalR(u)ightarrowmi{n}_u$ ; where the functional $calD$ is a data fidelity term also depending on some input data $f$ and measuring the deviation of $Ku$ from such and $calR$ is a regularization functional. Moreover $K$ is a (often linear) forward operator modeling the dependence of data on an underlying image, and $alpha$ is a positive regularization parameter. While $calD$ is often smooth and (strictly) convex, the current practice almost exclusively uses nonsmooth regularization functionals. The majority of successful techniques is using nonsmooth and convex functionals like the total variation and generalizations thereof or $el{l}_1$-norms of coefficients arising from scalar products with some frame system. The efficient solution of such variational problems in imaging demands for appropriate algorithms. Taking into account the specific structure as a sum of two very different terms to be minimized, splitting algorithms are a quite canonical choice. Consequently this field has revived the interest in techniques like operator splittings or augmented Lagrangians. Here we shall provide an overview of methods currently developed and recent results as well as some computational studies providing a comparison of different methods and also illustrating their success in applications.

Full work available at URL: https://arxiv.org/abs/1412.4237

zbMATH Keywords

numerical examplesalgorithmimage processingoperator splitting methodprimal-dual methodsLagrangian methodsnon-smooth variational models

Mathematics Subject Classification ID

Numerical aspects of computer graphics, image analysis, and computational geometry (65D18)

Cited In (39)

• A variable projection method for large-scale inverse problems with \(\ell^1\) regularization
• Backtracking Strategies for Accelerated Descent Methods with Smooth Composite Objectives
• Convergence Results for Primal-Dual Algorithms in the Presence of Adjoint Mismatch
• Iterative multiplicative filters for data labeling
• Automated parameter selection in the ${L}^{1} \mbox{-} {L}^{2}$-TV model for removing Gaussian plus impulse noise
• A Penalty Relaxation Method for Image Processing Using Euler's Elastica Model
• Accelerated alternating minimization algorithm for Poisson noisy image recovery
• An Operator-Splitting Method for the Gaussian Curvature Regularization Model with Applications to Surface Smoothing and Imaging
• Proximal operator of quotient functions with application to a feasibility problem in query optimization
• Shared Prior Learning of Energy-Based Models for Image Reconstruction
• Reconstruction Methods in THz Single-Pixel Imaging
• Strain analysis by a total generalized variation regularized optical flow model
• Implicit regularization with strongly convex bias: Stability and acceleration
• PDE-based group equivariant convolutional neural networks
• Domain decomposition for non-smooth (in particular TV) minimization
• A new algorithm framework for image inpainting in transform domain
• A fast algorithm for deconvolution and Poisson noise removal
• Regularization in Banach spaces with respect to the Bregman distance
• Variable Metric Forward-Backward Algorithm for Composite Minimization Problems
• Combined first and second order variational approaches for image processing
• Robust PCA via regularized \textsc{Reaper} with a matrix-free proximal algorithm
• On the convergence of recursive SURE for total variation minimization
• Transport between RGB images motivated by dynamic optimal transport
• Convolutional proximal neural networks and plug-and-play algorithms
• Extended randomized Kaczmarz method for sparse least squares and impulsive noise problems
• A Variational Model for Color Assignment
• Convergence of functions and their Moreau envelopes on Hadamard spaces
• Overlapping Domain Decomposition Methods for Total Variation Denoising
• Priors with coupled first and second order differences for manifold-valued image processing
• Denoising of sphere- and SO(3)-valued data by relaxed Tikhonov regularization
• Risk estimators for choosing regularization parameters in ill-posed problems -- properties and limitations
• Nonsmooth convex optimization for structured illumination microscopy image reconstruction
• Parseval proximal neural networks
• An Image Registration Model in Electron Backscatter Diffraction
• Representation and reconstruction of covariance operators in linear inverse problems
• Enhancing Compressed Sensing 4D Photoacoustic Tomography by Simultaneous Motion Estimation
• Convergence rate of inertial proximal algorithms with general extrapolation and proximal coefficients
• Curvature Regularized Surface Reconstruction from Point Clouds
• Optimal multivariate Gaussian fitting with applications to PSF modeling in two-photon microscopy imaging

This page was built for publication: First Order Algorithms in Variational Image Processing

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5350482)