Abstract
The dual-front scheme is a powerful curve evolution tool for active contours and image segmentation, which has proven its capability in dealing with various segmentation tasks. In its basic formulation, a contour is represented by the interface of two adjacent Voronoi regions derived from the geodesic distance map which is the solution to an Eikonal equation. The original dual-front model [17] is based on isotropic metrics, and thus cannot take into account the asymmetric enhancements during curve evolution. In this paper, we propose a new asymmetric dual-front curve evolution model through an asymmetric Finsler geodesic metric, which is constructed in terms of the extended normal vector field of the current contour and the image data. The experimental results demonstrate the advantages of the proposed method in computational efficiency, robustness and accuracy when compared to the original isotropic dual-front model.
Original language | English |
---|---|
Publication status | Published - 6 Sept 2018 |
Event | 29th British Machine Vision Conference, BMVC 2018 - Newcastle, United Kingdom Duration: 3 Sept 2018 → 6 Sept 2018 |
Conference
Conference | 29th British Machine Vision Conference, BMVC 2018 |
---|---|
Country/Territory | United Kingdom |
City | Newcastle |
Period | 3/09/18 → 6/09/18 |
Keywords
- active contours
- image segmentation
- dual-front model