跳到主要导航 跳到搜索 跳到主要内容

C2FNet: Cross-Probabilistic Weak Supervision Learning for High-Resolution Land Cover Enhancement

  • Boaz Mwubahimana*
  • , Jianguo Yan*
  • , Dingruibo Miao
  • , Zhuohong Li
  • , Haonan Guo
  • , Le Ma
  • , Maurice Mugabowindekwe
  • , Swalpa Kumar Roy
  • , Xiao Huang
  • , Elias Nyandwi
  • , Tuyishimire Joseph
  • , Eric Habineza
  • , Fidele Mwizerwa
  • , Hafashimana Anthanase
  • , Gaspard Rwanyiziri
  • *此作品的通讯作者
  • Duke University
  • University of Copenhagen
  • Alipurduar Government Engineering and Management College
  • Emory University
  • University of Rwanda

科研成果: 期刊稿件 › 论文 › 同行评审

8 引用 (Scopus)

摘要

Autonomous large-scale high-resolution land cover (HRLC) mapping remains a major challenge in remote sensing due to the scarcity of reliable training data and resolution mismatches between available labels and input of massive of emerging imagery. Existing global land cover products often suffer from coarse spatial resolution and label noise, limiting their utility for fine-scale urban analysis and environmental monitoring. This article presents C2FNet, a novel Coarse-to-Fine Network designed to generate HRLC maps from noisy, coarse-resolution labels using a weak supervision strategy of the cross-probability. The C2FNet consists of three key modules: 1) edge resolution refinement backbones (ERRBs), which preserve spatial detail via multiscale feature extraction through parallel convolutional branches; 2) unsupervised dynamic shuffle and diagonal annotation (UDSDA), which enhances training reliability by identifying confident regions through spatial-consistency analysis and confidence estimation; and 3) a contrasting self-supervised loss (C2F-Loss) that integrates cross-entropy and cosine similarity terms to mitigate supervision noise and resolution gaps. Evaluations of three benchmark datasets that encompass diverse urban and rural landscapes show that C2FNet achieves state-of-the-art (SoA) performance, with 80.01% overall accuracy (OA) and a Cohen’s kappa score of 0.7567, outperforming SoA models with weak supervision. © 2025 IEEE. All rights reserved.
源语言英语
期刊论文编号5654930
页数30
期刊IEEE Transactions on Geoscience and Remote Sensing
卷63
DOI
出版状态已出版 - 1月 2025

基金资助

European Space Agency, Number: -, Acronym: USGS, U.S. Geological Survey, Number: -, Acronym: FDCT, Fundo para o Desenvolvimento das Ciências e da Tecnologia, Number: -, Acronym: LIESMARS, State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Number: -, Acronym: WHU, Wuhan University, Number: 42241116, Acronym: NNSFC, National Natural Science Foundation of China, Number: -, Acronym: NNSFC, National Natural Science Foundation of China, Number: 2022YFF0503202, Acronym: NKPs, National Key Research and Development Program of China, Number: -, Acronym: NKPs, National Key Research and Development Program of China, Number: SKL-LPS(MUST)-2021-2023, Acronym: -, - (ESA)

关键词

  • Coarse-to-fine networks (C2FNets)
  • cross-resolution learning
  • deep neural networks
  • Earth observation
  • land cover mapping
  • probabilistic supervision
  • remote sensing
  • weakly supervised learning (WSL)

收录

  • SCIE
  • EI
  • Scopus

学术指纹

探究 'C2FNet: Cross-Probabilistic Weak Supervision Learning for High-Resolution Land Cover Enhancement' 的科研主题。它们共同构成独一无二的学术指纹。

引用此