Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
February 17, 2026Open Access

A Two-Stage Concrete Crack Segmentation Method Based on the Improved YOLOv11 and Segment Anything Model

View Full Paper
Ask AI
Bookmark
Share

Authors

RZRu ZhangCGChaodong GuanYFYi Fang

Discussion

Loading...

Member takes

Overview

This method identifies concrete cracks accurately in structures, suggesting effective monitoring for safety.

Key Points

  • The research aims to develop a two-stage method for accurate concrete crack segmentation to assess structural integrity.
  • Implemented an improved YOLOv11 model incorporating Multi-scale Edge Information Enhancement, Efficient-Detection, and P2-Level Feature Integration.
  • Employed Segment Anything Model (SAM) for precise crack segmentation using automatic prompts.
  • Introduced a mask re-input strategy for enhanced segmentation performance under variable lighting conditions.
  • Achieved average Accuracy of 95.98%, precision of 92.60%, and Intersection over Union (IoU) of 0.77 in initial segmentation.
  • Maintained strong segmentation with an Average Accuracy of 92.38%, precision of 85.70%, and IoU of 0.64 in follow-up segmentation.

Cite This Study

Zhang et al. (2026) studied this question.

synapsesocial.com/papers/6994058c4e9c9e835dfd6776https://doi.org/10.3390/buildings16040794
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Automated Detection and Segmentation of Cracks in Urban Underground Structures Based on YOLOv8-SAM22026
  2. 2Crack-SAM: Crack Segmentation Using a Foundation Model2024
  3. 3Engineering-oriented automated segmentation and quantitative analysis of building structural cracks: A multi-mechanism optimized YOLOv11-seg approach2026
  4. 4Concrete Surface Crack Detection Algorithm Based on Improved YOLOv82024 · 58 citations
  5. 5Segment anything model-based crack segmentation using low-rank adaption fine-tuning2024 · 2 citations