.png){kind=icon}
.png){kind=icon}
Return
TF-RefineNet: A deep time–frequency refinement network and its application to tacho-less order tracking
.png){kind=icon}
PRE
AI
M
王
T
L
Y
汤
DOI：10.1016/j.ymssp.2026.114845.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A TF-ridge-aware mask estimation paradigm is proposed for deep time–frequency refinement. • An ideal binary time–frequency mask and Dice loss generate clean TFRs with only ridges retained. • A local–global cross-scale fusion module captures local ridge details and global ridge trends. • Ridge-clear TFRs support accurate IF extraction and tacho-less order tracking.
Keywords:
Deep time–frequency analysis
Instantaneous frequency
Time-frequency ridge-aware
Ideal binary time–frequency mask
Tacho-less order tracking
Journal
IF:
8.9
Papers:
1.2W
Citations:
6.6W
Organization
Cited Papers