Header

UZH-Logo

Maintenance Infos

Coupling Image Restoration and Segmentation: A Generalized Linear Model/Bregman Perspective


Paul, Grégory; Cardinale, Janick; Sbalzarini, Ivo F (2013). Coupling Image Restoration and Segmentation: A Generalized Linear Model/Bregman Perspective. International Journal of Computer Vision, 104(1):69-93.

Abstract

We introduce a new class of data-fitting energies that couple image segmentation with image restoration. These functionals model the image intensity using the statistical framework of generalized linear models. By duality, we establish an information-theoretic interpretation using Bregman divergences. We demonstrate how this formulation couples in a principled way image restoration tasks such as denoising, deblurring (deconvolution), and inpainting with segmentation. We present an alternating minimization algorithm to solve the resulting composite photometric/geometric inverse problem.We use Fisher scoring to solve the photometric problem and to provide asymptotic uncertainty estimates. We derive the shape gradient of our data-fitting energy and investigate convex relaxation for the geometric problem. We introduce a new alternating split- Bregman strategy to solve the resulting convex problem and present experiments and comparisons on both synthetic and real-world images.

Abstract

We introduce a new class of data-fitting energies that couple image segmentation with image restoration. These functionals model the image intensity using the statistical framework of generalized linear models. By duality, we establish an information-theoretic interpretation using Bregman divergences. We demonstrate how this formulation couples in a principled way image restoration tasks such as denoising, deblurring (deconvolution), and inpainting with segmentation. We present an alternating minimization algorithm to solve the resulting composite photometric/geometric inverse problem.We use Fisher scoring to solve the photometric problem and to provide asymptotic uncertainty estimates. We derive the shape gradient of our data-fitting energy and investigate convex relaxation for the geometric problem. We introduce a new alternating split- Bregman strategy to solve the resulting convex problem and present experiments and comparisons on both synthetic and real-world images.

Statistics

Citations

18 citations in Web of Science®
21 citations in Scopus®
Google Scholar™

Altmetrics

Downloads

123 downloads since deposited on 05 Jul 2013
18 downloads since 12 months
Detailed statistics

Additional indexing

Item Type:Journal Article, refereed, original work
Communities & Collections:Special Collections > SystemsX.ch
Special Collections > SystemsX.ch > Research, Technology and Development Projects > LipidX
Special Collections > SystemsX.ch > Research, Technology and Development Projects > WingX
Dewey Decimal Classification:570 Life sciences; biology
Language:English
Date:2013
Deposited On:05 Jul 2013 10:16
Last Modified:05 Apr 2016 16:51
Publisher:Springer New York LLC
ISSN:0920-5691
Publisher DOI:https://doi.org/10.1007/s11263-013-0615-2

Download

Preview Icon on Download
Preview
Filetype: PDF
Size: 11MB
View at publisher