.png){kind=icon}
.png){kind=icon}
Return
Semi-supervised deep learning method for abnormal detection based on ship trajectory
.png){kind=icon}
PRE
AI
DOI：10.1016/j.ress.2026.113264.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A semi-supervised method is proposed for abnormal ship trajectory detection. • Adaptive grid partitioning is designed for dense lake AIS trajectories. • Dynamic weighted joint loss balances reconstruction and classification. • A joint learning framework enhances detection under limited labels.
Keywords:
Abnormal detection
Semi-supervised deep learning
Adaptive grid partitioning
Transformer
AIS dataset
Journal
R
IF:
11
Papers:
733
Citations:
0