Enhancing SAM-based digital rock image segmentation via edge-semantics fusion

The Segment Anything Model (SAM) demonstrates strong segmentation capabilities. However, its application to digital rock images faces challenges from subtle transitions between matrix minerals and pore structures, as well as inherent heterogeneity, which result in mis-segmentation and discontinuitie...

Full description

Bibliographic Details

Published in:	Applied Computing and Geosciences
Main Authors:	Ziqiang Wang, Zhiyu Hou, Danping Cao
Format:	Article
Language:	English
Published:	Elsevier 2025-12-01
Subjects:	Digital rock physics Image segmentation Segment anything model
Online Access:	http://www.sciencedirect.com/science/article/pii/S2590197425000746