Abstract:In image processing, the RudinOsherFatemi (ROF) model [L. Rudin, S. Osher, and E. Fatemi, Phys. D, 60 (1992), pp. 259268] based on total variation (TV) minimization has proven to be very useful. So far many researchers have contributed to designing fast numerical schemes and overcoming the nondifferentiability of the model. Methods considered to be particularly efficient for the ROF model include the ChanGolubMulet (CGM) primal-dual method [T.F. Chan, G.H. Golub, and P. Mulet, SIAM J. Sci. Comput., 20 (1999), pp. 19641977], Chambolle's dual method [A. Chambolle, J. Math. Imaging Vis., 20 (2004), pp. 8997], the splitting and quadratic penalty-based method [Y. Wang, J. Yang, W. Yin, and Y. Zhang, SIAM J. Imaging Sci., 1 (2008), pp. 248272], and the split Bregman iteration [T. Goldstein and S. Osher, SIAM J. Imaging Sci., 2 (2009), pp. 323343], as well as the augmented Lagrangian method [X.C. Tai and C. Wu, Lecture Notes in Comput. Sci. 5567, Springer-Verlag, Berlin, 2009, pp. 502513]. In this paper, we first review the augmented Lagrangian method for the ROF model and then provide some convergence analysis and extensions to vectorial TV and high order models. All the algorithms and analysis will be presented in the discrete setting, which is much clearer for practical implementation than the continuous setting as in Tai and Wu, above. We also present, in the discrete setting, the connections between the augmented Lagrangian method, the dual methods, and the split Bregman iteration. Using our extensions and observations, we can easily figure out CGM and the split Bregman iteration for vectorial TV and high order models, which, to the best of our knowledge, have not been presented in the literature. Numerical examples demonstrate the efficiency and accuracy of our method, especially in the image deblurring case.

An Algorithm Based on Augmented Lagrangian Method for Generalized Gradient Vector Flow Computation.

Fast gradient vector flow computation based on augmented Lagrangian method

An augmented Lagrangian method for fast gradient vector flow computation

Efficient Generalized Fused Lasso and Its Applications.

Multigrid based fast implicit GVF method

Mean Shift Based Gradient Vector Flow for Image Segmentation

An Efficient Augmented Lagrangian Method with Semismooth Newton Solver for Total Generalized Variation.

Fast Gradient Computation for Gromov-Wasserstein Distance

BDF schemes for accelerated gradient flows in projection-free approximation of nonconvex constrained variational minimization

Accelerated Gradient and Skew-Symmetric Splitting Methods for a Class of Monotone Operator Equations

Augmented Lagrangian Method, Dual Methods, and Split Bregman Iteration for ROF, Vectorial TV, and High Order Models

A Fast Unsmoothed Aggregation Algebraic Multigrid Framework for the Large-Scale Simulation of Incompressible Flow.

Image Structure Adaptive Gradient Vector Flow for Active Contours

Fast Augmented Lagrangian Method for Image Smoothing with Hyper-Laplacian Gradient Prior.

Greedy Algorithm Based Deformable Simplex Meshes Using Gradient Vector Flow As External Energy

Generalized Gradient Vector Flow for Snakes: New Observations, Analysis, and Improvement

A modified variational method for large displacement optical flow

A Grid-Free Fluid Solver based on Gaussian Spatial Representation

Gradient Flows and Riemannian Structure in the Gromov-Wasserstein Geometry

GPU-Accelerated Flow Simulations on Unstructured Grids Using a Multi-colored Gauss-Seidel Method

Sigmoid gradient vector flow for medical image segmentation