Hyperbolic Prototype Guidance for Incremental SAR ATR
Despite the success of deep learning methods in synthetic aperture radar automatic target recognition (SAR ATR), they face significant challenges in open environments. Deep models suffer from catastrophic forgetting when learning new classes. Moreover, SAR images are frequently degraded by serious n...
Uloženo v:
| Hlavní autoři: | , , , , , |
|---|---|
| Médium: | Artigo |
| Jazyk: | Inglês |
| Vydáno: |
IEEE
2026-01-01
|
| Edice: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Témata: | |
| On-line přístup: | https://ieeexplore.ieee.org/document/11560891/ |
| Tagy: |
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|