Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
June 14, 2026Scientific ReportsOpen Access

Dynamic multi-scale fusion network for road damage detection in complex street views

View Full Paper
Ask AI
Bookmark
Share

Authors

HZHong ZhouHuaiyin Institute of TechnologyYQYiyang QinHuaiyin Normal UniversityJRJiahuan RenHuaiyin Normal University

Discussion

Loading...

Member takes

Implication

Randomized trial reveals improved road damage detection using a new multi-scale method in complex street views, indicating enhanced safety.

Key Points

  • This research aims to develop an advanced method for detecting road damage that addresses challenges in varying damage sizes and complex geometries.
  • Proposed a dynamic multi-scale fusion network (DMSFNet) to enhance feature extraction capabilities in road damage detection.
  • Introduced a group fusion block (GFB) for multi-scale hybrid convolution and a multi-branch feature fusion (MBFF) module to capture contextual features.
  • Conducted experiments using SVRDD, RDD2020, and USRDD datasets.
  • Achieved significant improvements in detection accuracy for various sizes of road damage against existing methods.
  • Demonstrated a favorable balance between detection accuracy and computational efficiency.

Cite This Study

Zhou et al. (2026) studied this question.

synapsesocial.com/papers/6a2e44e4b1cc60ccdea8a52fhttps://doi.org/10.1038/s41598-026-56949-w
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. 1Real-Time Dynamic Scale-Aware Fusion Detection Network: Take Road Damage Detection as an example2024
  2. 2YOLOX-RDD: A Method of Anchor-Free Road Damage Detection for Front-View Images2024 · 27 citations
  3. 3Automatic Road Damage Detection Based on Improved YOLO112026
  4. 4Automatic road damage recognition based on improved YOLOv11 with multi-scale feature extraction and fusion attention mechanism2025 · 10 citations
  5. 5Artificial Intelligence‐Driven Multi‐Class Road Damage Detection Using Attention‐Enhanced YOLOv8 and Weighted Boxes Fusion2026