A Variational Approach to Joint Denoising, Edge Detection and Motion Estimation

Telea, Alexandru; Preußer, Tobias; Garbe, Christoph S.; Droske, Marc; Rumpf, Martin

doi:10.1007/11861898_53

Cited by 10 publications

(4 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We refer to [1], [2], [5], and [6]. In the context of the optical flow, this method has been less frequently used, e.g., in [33] (however, with quite disappointing numerical results), [41], and [10].…”

Section: Notationmentioning

confidence: 99%

Detection of Intensity and Motion Edges within Optical Flow via Multidimensional Control

Brüne

Maurer

Wagner³

2009

SIAM J. Imaging Sci.

View full text Add to dashboard Cite

In this paper, we propose a new optimization approach for the simultaneous computation of optical flow and edge detection therein. Instead of using an Ambrosio-Tortorelli type energy functional, we reformulate the optical flow problem as a multidimensional control problem. The optimal control problem is solved by discretization methods and large-scale optimization techniques. The edge detector can be immediately built from the control variables. We provide three series of numerical examples. The first shows that the mere presence of a gradient restriction has a regularizing effect, while the second demonstrates how to balance the regularizing effects of a term within the objective and the control restriction. The third series of numerical results is concerned with the direct evaluation of a TV-regularization term by introduction of control variables with sign restrictions.

show abstract

Section: Notationmentioning

confidence: 99%

Detection of Intensity and Motion Edges within Optical Flow via Multidimensional Control

Brüne

Maurer

Wagner³

2009

SIAM J. Imaging Sci.

View full text Add to dashboard Cite

show abstract

“…A traditional approach integrates an image reconstruction algorithm for removing noise and artifacts as a pre-processing step to the registration task [29]. Further improvements can be achieved by developing a joint framework that alternates between image reconstruction and registration [12,16,28,32]. The most widely used image reconstruction algorithms are based on optimization of an objective function that includes a regularization term for mitigating noise.…”

Section: Introductionmentioning

confidence: 99%

Plug-and-Play Priors for Reconstruction-Based Placental Image Registration

Xing

Kamilov

et al. 2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

This paper presents a novel deformable registration framework, leveraging an image prior specified through a denoising function, for severely noise-corrupted placental images. Recent work on plug-andplay (PnP) priors has shown the state-of-the-art performance of reconstruction algorithms under such priors in a range of imaging applications. Integration of powerful image denoisers into advanced registration methods provides our model with a flexibility to accommodate datasets that have low signal-to-noise ratios (SNRs). We demonstrate the performance of our method under a wide variety of denoising models in the context of diffeomorphic image registration. Experimental results show that our model substantially improves the accuracy of spatial alignment in applications of 3D in-utero diffusion-weighted MR images (DW-MRI) that suffer from low SNR and large spatial transformations.

show abstract

“…Simultaneous approaches [11,12] also involve motion segmentation rather than displacement estimation. Recently [12,13,14], edge detection is used jointly with the image de-noising and motion segmentation processes.…”

Section: Introductionmentioning

confidence: 99%

Simultaneous image de-noising and registration using graph cuts: Application to corrupted medical images

Lombaert

Cheriet

2012

2012 11th International Conference on Information Science, Signal Processing and Their Applications (ISSPA)

View full text Add to dashboard Cite

The success of medical image de-noising often relies on the image quality. If the image is severely degraded, information can be permanently lost. The de-noising or restoration process rarely use any other external information such as valuable data from additional images, for instance from a follow-up study or within an image sequence. Several optimization methods exist, among them the Graph Cuts method is efficient in a global optimum sense. We show that Graph Cuts can be used to solve simultaneously image de-noising and image correspondence. Both of these problems have been previously solved with Graph Cuts, but always as separate processes. In this paper, we combine them in the same formulation, and we show an application where images, initially unusable, can be recovered rather than being reacquired at a high risk (e.g., avoiding new radiation in medical scans).

show abstract

A Variational Approach to Joint Denoising, Edge Detection and Motion Estimation

Cited by 10 publications

References 24 publications

Detection of Intensity and Motion Edges within Optical Flow via Multidimensional Control

Detection of Intensity and Motion Edges within Optical Flow via Multidimensional Control

Plug-and-Play Priors for Reconstruction-Based Placental Image Registration

Simultaneous image de-noising and registration using graph cuts: Application to corrupted medical images

Contact Info

Product

Resources

About