Unique Perturbation Methods Exploitation for Semi-Supervised Remote Sensing Image Semantic Segmentation

Guardado en:
Detalles Bibliográficos
Publicado en:ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences vol. X-G-2025 (2025), p. 1085
Autor principal: Zhou, Liang
Otros Autores: Duan, Keyi, Dai, Jinkun, Wu, Xiaodan, Ge, Xuming, Li, Xiaojun, Ye, Yuanxin
Publicado:
Copernicus GmbH
Materias:
Acceso en línea:Citation/Abstract
Full Text - PDF
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!

MARC

LEADER 00000nab a2200000uu 4500
001 3243894528
003 UK-CbPIL
022 |a 2194-9042 
022 |a 2194-9050 
024 7 |a 10.5194/isprs-annals-X-G-2025-1085-2025  |2 doi 
035 |a 3243894528 
045 2 |b d20250101  |b d20251231 
084 |a 263032  |2 nlm 
100 1 |a Zhou, Liang  |u Faculty of Geosciences and Engineering, Southwest Jiaotong University, Chengdu, Sichuan, China; Faculty of Geosciences and Engineering, Southwest Jiaotong University, Chengdu, Sichuan, China 
245 1 |a Unique Perturbation Methods Exploitation for Semi-Supervised Remote Sensing Image Semantic Segmentation 
260 |b Copernicus GmbH  |c 2025 
513 |a Journal Article 
520 3 |a Deep learning has significantly improved the accuracy of remote sensing semantic segmentation, yet its effectiveness is often constrained by the limited availability of annotated training samples. Semi-supervised learning (SSL) addresses this challenge by utilizing abundant unlabeled data, reducing dependence on manual annotations. However, current consistency regularization-based SSL methods, primarily developed for natural images, struggle to produce adequate perturbation diversity for robust model training in remote sensing image segmentation. In this work, we propose FusionMatch, a novel SSL framework featuring two perturbation mechanisms - NIRPerb and PSPerb - specifically designed for remote sensing imagery. NIRPerb utilizes near-infrared spectral data to enhance perturbation diversity. PSPerb adopts differentiated pan-sharpening fusion strategies to expand the perturbation space. Extensive experiments on both a building extraction dataset and a multi-class dataset demonstrate that FusionMatch outperforms state-of-the-art SSL methods in segmentation accuracy and robustness. 
653 |a Regularization 
653 |a Datasets 
653 |a Image segmentation 
653 |a Near infrared radiation 
653 |a Remote sensing 
653 |a Infrared imagery 
653 |a Semantic segmentation 
653 |a Training 
653 |a Semi-supervised learning 
653 |a Deep learning 
653 |a Perturbation methods 
653 |a Infrared spectra 
653 |a Peer review 
653 |a Photogrammetry 
653 |a Supervision 
653 |a Regularization methods 
653 |a Architecture 
653 |a Annotations 
653 |a Semantics 
653 |a Information science 
653 |a Environmental 
700 1 |a Duan, Keyi  |u Faculty of Geosciences and Engineering, Southwest Jiaotong University, Chengdu, Sichuan, China; Faculty of Geosciences and Engineering, Southwest Jiaotong University, Chengdu, Sichuan, China 
700 1 |a Dai, Jinkun  |u Faculty of Geosciences and Engineering, Southwest Jiaotong University, Chengdu, Sichuan, China; Faculty of Geosciences and Engineering, Southwest Jiaotong University, Chengdu, Sichuan, China 
700 1 |a Wu, Xiaodan  |u Faculty of Geosciences and Engineering, Southwest Jiaotong University, Chengdu, Sichuan, China; Faculty of Geosciences and Engineering, Southwest Jiaotong University, Chengdu, Sichuan, China 
700 1 |a Ge, Xuming  |u Faculty of Geosciences and Engineering, Southwest Jiaotong University, Chengdu, Sichuan, China; Faculty of Geosciences and Engineering, Southwest Jiaotong University, Chengdu, Sichuan, China 
700 1 |a Li, Xiaojun  |u Faculty of Geosciences and Engineering, Southwest Jiaotong University, Chengdu, Sichuan, China; Faculty of Geosciences and Engineering, Southwest Jiaotong University, Chengdu, Sichuan, China 
700 1 |a Ye, Yuanxin  |u Faculty of Geosciences and Engineering, Southwest Jiaotong University, Chengdu, Sichuan, China; Faculty of Geosciences and Engineering, Southwest Jiaotong University, Chengdu, Sichuan, China 
773 0 |t ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences  |g vol. X-G-2025 (2025), p. 1085 
786 0 |d ProQuest  |t Engineering Database 
856 4 1 |3 Citation/Abstract  |u https://www.proquest.com/docview/3243894528/abstract/embedded/CH9WPLCLQHQD1J4S?source=fedsrch 
856 4 0 |3 Full Text - PDF  |u https://www.proquest.com/docview/3243894528/fulltextPDF/embedded/CH9WPLCLQHQD1J4S?source=fedsrch