An extended Perona-Malik model based on probabilistic models

DOI10.1007/S10851-017-0746-0zbMATH1387.94020OpenAlexW2583099386MaRDI QIDQ1702003FDOQ1702003

Authors L. M. Mescheder, Dirk A. Lorenz

Publication date 27 February 2018

Published in Journal of Mathematical Imaging and Vision (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1612.06176

zbMATH Keywords

mean field approximationprobabilistic modelsGaussian scale mixturesPerona-Malik denoising

Mathematics Subject Classification ID

Image analysis in multivariate analysis (62H35) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08)

Abstract: The Perona-Malik model has been very successful at restoring images from noisy input. In this paper, we reinterpret the Perona-Malik model in the language of Gaussian scale mixtures and derive some extensions of the model. Specifically, we show that the expectation-maximization (EM) algorithm applied to Gaussian scale mixtures leads to the lagged-diffusivity algorithm for computing stationary points of the Perona-Malik diffusion equations. Moreover, we show how mean field approximations to these Gaussian scale mixtures lead to a modification of the lagged-diffusivity algorithm that better captures the uncertainties in the restoration. Since this modification can be hard to compute in practice we propose relaxations to the mean field objective to make the algorithm computationally feasible. Our numerical experiments show that this modified lagged-diffusivity algorithm often performs better at restoring textured areas and fuzzy edges than the unmodified algorithm. As a second application of the Gaussian scale mixture framework, we show how an efficient sampling procedure can be obtained for the probabilistic model, making the computation of the conditional mean and other expectations algorithmically feasible. Again, the resulting algorithm has a strong resemblance to the lagged-diffusivity algorithm. Finally, we show that a probabilistic version of the Mumford-Shah segementation model can be obtained in the same framework with a discrete edge-prior.

Recommendations

• An approach to hybrid probabilistic models
• Bayesian probabilistic extensions of a deterministic model
• On a Probabilistic Bernstein Model
• scientific article; zbMATH DE number 4071261
• Publication:3475262
• Modal Interval Probability: Application to Bonus-Malus Systems

Cites work

• scientific article; zbMATH DE number 1194487 (Why is no real title available?)
• scientific article; zbMATH DE number 3567782 (Why is no real title available?)
• scientific article; zbMATH DE number 1111402 (Why is no real title available?)
• scientific article; zbMATH DE number 3442988 (Why is no real title available?)
• scientific article; zbMATH DE number 3088518 (Why is no real title available?)
• Augmented Tikhonov regularization
• Gaussian Markov Random Fields
• Graphical models, exponential families, and variational inference
• Image denoising using scale mixtures of gaussians in the wavelet domain
• Iterative Methods for Total Variation Denoising
• Julia: a fresh approach to numerical computing
• MCMC-based image reconstruction with uncertainty quantification
• Machine learning. A probabilistic perspective
• Mumford–Shah Functional as Γ-Limit of Discrete Perona–Malik Energies
• Nonlinear total variation based noise removal algorithms
• On the Convergence of the Lagged Diffusivity Fixed Point Method in Total Variation Image Restoration
• Optimal approximations by piecewise smooth functions and associated variational problems
• Pattern recognition and machine learning.
• Statistical and computational inverse problems.
• Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images
• Variational methods in imaging

Cited in

(1)

• Perona-Malik model with self-adjusting shape-defining constant

Describes a project that uses

Uses Software

• GMRFLib
• Julia
• PMTK
• PRMLT

This page was built for publication: An extended Perona-Malik model based on probabilistic models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1702003)