.png){kind=icon}
.png){kind=icon}
Return
Fine-grained vehicle damage classification based on applicability network
.png){kind=icon}
PRE
AI
DOI：10.1016/j.asoc.2026.115076.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A universally applicable network for damage and defect classification, FGDANet, is proposed. • Independent feature extraction (IR Layer) and regional learning modules (RL Block) are introduced. • An adaptive global feature attention algorithm (GFS Module) is proposed. • The generalized feature constraint loss (GFC Loss) is introduced. • The network demonstrates excellent performance in both vehicle damage classification and steel surface defect classification tasks.
Keywords:
FGDANet
IR Layer
RL Block
GFS Module
GFC Loss
Journal
IF:
6.6
Papers:
1.4W
Citations:
4.8W
Organization
No organization information available