A Fast Anisotropic Mumford-Shah Functional Based Segmentation

Authors:
J. F. Garamendi;N. Malpica;E. Schiavi
Affiliations:
Universidad Rey Juan Carlos, Móstoles, Madrid, Spain;Universidad Rey Juan Carlos, Móstoles, Madrid, Spain;Universidad Rey Juan Carlos, Móstoles, Madrid, Spain
Venue:
IbPRIA '09 Proceedings of the 4th Iberian Conference on Pattern Recognition and Image Analysis
Year:
2009

Citing 8
Cited 0

Nonlinear total variation based noise removal algorithms

Proceedings of the eleventh annual international conference of the Center for Nonlinear Studies on Experimental mathematics : computational issues in nonlinear science: computational issues in nonlinear science
Multigrid

Multigrid
Geometric models for active contours

ICIP '95 Proceedings of the 1995 International Conference on Image Processing (Vol. 3)-Volume 3 - Volume 3
Geometric Level Set Methods in Imaging,Vision,and Graphics

Geometric Level Set Methods in Imaging,Vision,and Graphics
An Algorithm for Total Variation Minimization and Applications

Journal of Mathematical Imaging and Vision
Automatic Segmentation of the Liver in CT Using Level Sets Without Edges

IbPRIA '07 Proceedings of the 3rd Iberian conference on Pattern Recognition and Image Analysis, Part I
Reformulating and Optimizing the Mumford-Shah Functional on a Graph -- A Faster, Lower Energy Solution

ECCV '08 Proceedings of the 10th European Conference on Computer Vision: Part I
Active contours without edges

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Digital (binary) image segmentation is a critical step in most image processing protocols, especially in medical imaging where accurate and fast segmentation and classification are a challenging issue. In this paper we present a fast relaxation algorithm to minimize an anistropic Mumford-Shah energy functional for piecewise constant approximation of corrupted data. The algorithm is tested with synthetic phantoms and some CT images of the abdomen. Our results are finally compared with manual segmentations in order to validate the proposed model.