.png){kind=icon}
.png){kind=icon}
Return
SDANet: A structure-detail aware network for image restoration in enclosed building spaces under fire-smoke conditions
.png){kind=icon}
OA
AI
W
M
Y
Y
C
J
C
X
DOI：10.1016/j.dibe.2026.101010.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• SFAE jointly models channel–spatial responses for non-uniform smoke perception. • MFDE uses Haar wavelet decomposition to preserve high-frequency structural details. • CNPF adaptively fuses shallow detail and deep semantic features across scales. • A Unity3D-based paired fire-smoke dataset is constructed for enclosed building spaces. • SDANet improves PSNR by 4.57 dB and SSIM by 0.07 over the best-performing baseline.
Keywords:
Image restoration
Structure-detail aware
Enclosed building spaces
Fire-smoke conditions
Smoke removal
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
8.2
Papers:
985
Citations:
3.3K
Organization
Cited Papers