Finsler Geodesics Evolution Model for Region based Active Contours

Da Chen, Jean-Marie Mirebeau and Laurent Cohen

Abstract

In this paper, we present a new deformable model for image segmentation by building the connection between the region based active contours energy and the geodesic energy via a Finsler metric. Basically, we solve the region based active contours energy minimization problem under a nonlinear Eikonal equation framework with respect to a Finsler metric. By sampling a set of control points from the closed active contour in clockwise order, the active contours evolution problem is turned to find a collection of minimal curves joining all the control points. These minimal curves can be obtained by solving the Finsler metric based Eikonal equation. Benefiting from the well established numerical solver of the Eikonal PDE, named fast marching algorithm, our model can obtain a fast numerical solution.

Session

Posters 1

Files

Extended Abstract (PDF, 116K)
Paper (PDF, 2M)

DOI

10.5244/C.30.22
https://dx.doi.org/10.5244/C.30.22

Citation

Bibtex

        @inproceedings{BMVC2016_22,
        	title={Finsler Geodesics Evolution Model for Region based Active Contours},
        	author={Da Chen, Jean-Marie Mirebeau and Laurent Cohen},
        	year={2016},
        	month={September},
        	pages={22.1-22.12},
        	articleno={22},
        	numpages={12},
        	booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
        	publisher={BMVA Press},
        	editor={Richard C. Wilson, Edwin R. Hancock and William A. P. Smith},
        	doi={10.5244/C.30.22},
        	isbn={1-901725-59-6},
        	url={https://dx.doi.org/10.5244/C.30.22}
        }