.png){kind=icon}
.png){kind=icon}
Return
A skin lesion segmentation algorithm based on multi-scale martingale processes
.png){kind=icon}
PRE
AI
DOI：10.1016/j.displa.2026.103414.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A multi-scale texture martingale framework is introduced for robust dermatological image segmentation. • A new Markov random field energy function incorporating martingale distance and gradient penalties is proposed. • The proposed method achieves accurate and smooth lesion boundary delineation under challenging conditions. Extensive experiments on public datasets validate improved accuracy and texture boundary preservation.
Keywords:
skin lesion segmentation
multi-scale texture
martingale framework
Markov random field
image segmentation