A level set approach for shape recovery of open contours

Authors:
Min Li;Chandra Kambhamettu;Maureen Stone
Affiliations:
Department of Computer & Information Sciences, University of Delaware, Newark, DE;Department of Computer & Information Sciences, University of Delaware, Newark, DE;Dept of Biomedical Sciences and Orthodontics, Vocal Tract Visualization Lab, University of Maryland Dental School, Baltimore, MD
Venue:
ACCV'06 Proceedings of the 7th Asian conference on Computer Vision - Volume Part I
Year:
2006

Citing 7
Cited 0

Shape Modeling with Front Propagation: A Level Set Approach

IEEE Transactions on Pattern Analysis and Machine Intelligence
Geodesic Active Contours

International Journal of Computer Vision
Spiral crystal growth

Physica D
Shape Priors for Level Set Representations

ECCV '02 Proceedings of the 7th European Conference on Computer Vision-Part II
Reconstructing open surfaces from unorganized data points

CVPR'04 Proceedings of the 2004 IEEE computer society conference on Computer vision and pattern recognition
A topology preserving level set method for geometric deformable models

IEEE Transactions on Pattern Analysis and Machine Intelligence
Active contours without edges

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, a geometric deformable model for shape recovery of open contours in noisy images is presented. We use two level set functions to model the open contour and find the end points of the open contour as the intersection of the two level set functions. The evolutions of both level set functions do not depend on the gradient of the images, as in the classical geometric deformable models, but are decided by a region-based ”band velocity”. The ”band velocity” is different from region information introduced by other deformable models which can only be used to find the closed contours in images, it is designed for evolutions of both closed and open contours and particularly unique for contours which are open and do not enclose any region. Prior shape information is also integrated into the contour evolution process, which prevents two level set functions from intersecting at other places than at the contour end points. With the described method open contours can be recovered from noisy images. Successful experiments on several data sets are presented in this paper.