Calibrated Focal Loss for Semantic Labeling of High-Resolution Remote Sensing Images

Calibrated Focal Loss for Semantic Labeling of High-Resolution Remote Sensing Images

Currently, the most advanced high-resolution remote sensing image (HRRSI) semantic labeling methods rely on deep neural networks. However, HRRSIs naturally have a serious class imbalance problem, which is not yet well solved by the current method. The cross-entropy loss is often used to guide the tr...

Descripció completa

Dades bibliogràfiques
Autors principals: Haiwei Bai, Jian Cheng, Yanzhou Su, Siyu Liu, Xin Liu
Format: Article
Idioma:English
Publicat: IEEE 2022-01-01
Col·lecció:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Matèries:
Calibrated focal loss (CFL)
class imbalance
focal loss (FL)
high-resolution remote sensing images (HRRSIs)
prediction confusion map (PCM)
semantic labeling
Accés en línia:https://ieeexplore.ieee.org/document/9854049/

Internet

https://ieeexplore.ieee.org/document/9854049/

Ítems similars