DivLS: A Divergence-Driven Level Set Model with Regional Fitting for Image Segmentation
Journal of Imaging · 14.11.2026 · 12 (11) · 576
DOI: 10.3390/jimaging12110576 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Imaging, online erschienen am 14.11.2026. Verfasst von Ming Dai und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Guangdong Basic and Applied Basic Research Foundation.
Abstract (Originalsprache)
In edge-based level set segmentation, a non-negative edge-stopping function modulates the strength of a constant area force but cannot adapt the direction of this contribution to the local contour position. This paper presents DivLS, a divergence-driven level set model with regional fitting. A hyperbolic-tangent weight modulates the magnitude of the Gaussian-smoothed image gradient while preserving its direction. The outward flux of the resulting field is incorporated into an energy with regional fitting, contour length and double-well distance regularization. The resulting evolution combines the field divergence with foreground–background fitting costs to guide contour motion. The weighting response is analyzed, and a locally planar bright-foreground transition is used to explain the action of the image force on contours displaced to either side of the boundary. Experimental results on simulated MR images show favorable boundary localization and regional coverage. Ablations demonstrate complementary contributions of the flux and regional-fitting terms to boundary localization and foreground coverage, respectively.
Förderung
- Guangdong Basic and Applied Basic Research Foundation
Zitieren
APA
Dai, M., Deng, Y., & Chen, R. (2026). DivLS: A Divergence-Driven Level Set Model with Regional Fitting for Image Segmentation. Journal of Imaging, 12(11), 576. https://doi.org/10.3390/jimaging12110576
Vancouver
Dai M, Deng Y, Chen R. DivLS: A Divergence-Driven Level Set Model with Regional Fitting for Image Segmentation. Journal of Imaging. 2026;12(11):576. doi:10.3390/jimaging12110576
BibTeX
@article{Dai2026,
author = {Dai, Ming and Deng, Yangfang and Chen, Ruyun},
title = {DivLS: A Divergence-Driven Level Set Model with Regional Fitting for Image Segmentation},
journal = {Journal of Imaging},
year = {2026},
volume = {12},
number = {11},
pages = {576},
doi = {10.3390/jimaging12110576}
}
EUVIMED