HB-Net: Hybrid Label and Boundary-Optimized Network for Weakly Supervised Raft Aquaculture Areas Extraction in Remote Sensing Images

Accurate extraction of raft aquaculture areas from remote sensing imagery is essential for sustainable marine resource management. Conventional deep learning approaches rely heavily on large-scale pixelwise annotations, which are costly and impractical for large-area monitoring. To alleviate this bo...

ver descrição completa

Detalhes bibliográficos

Principais autores:	Binge Cui, Faqi Wang, Mengting Liu, Jun Xing, Haojie Li, Ling Huang
Formato:	Artigo
Idioma:	Inglês
Publicado em:	IEEE 2026-01-01
coleção:	IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Assuntos:	Boundary refinement (BR) geometric prior hybrid labeling scribble annotation weakly supervised learning
Acesso em linha:	https://ieeexplore.ieee.org/document/11258899/
Tags:	Adicionar Tag Sem tags, seja o primeiro a adicionar uma tag!