Unsupervised Multi-Scale Hybrid Feature Extraction Network for Semantic Segmentation of High-Resolution Remote Sensing Images
Remote Sensing · 11.10.2024 · 16 (20) · 3774
DOI: 10.3390/rs16203774 ↗Kurz gefasst
Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 11.10.2024. Verfasst von Wanying Song und 4 weiteren Autor:innen. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Foundation of China, Outstanding Youth Science Fund of Xi’an University of Science and Technology, Scientific Research Plan Projects of Shaanxi Education Department.
Abstract (Originalsprache)
Generating pixel-level annotations for semantic segmentation tasks of high-resolution remote sensing images is both time-consuming and labor-intensive, which has led to increased interest in unsupervised methods. Therefore, in this paper, we propose an unsupervised multi-scale hybrid feature extraction network based on the CNN-Transformer architecture, referred to as MSHFE-Net. The MSHFE-Net consists of three main modules: a Multi-Scale Pixel-Guided CNN Encoder, a Multi-Scale Aggregation Transformer Encoder, and a Parallel Attention Fusion Module. The Multi-Scale Pixel-Guided CNN Encoder is designed for multi-scale, fine-grained feature extraction in unsupervised tasks, efficiently recovering local spatial information in images. Meanwhile, the Multi-Scale Aggregation Transformer Encoder introduces a multi-scale aggregation module, which further enhances the unsupervised acquisition of multi-scale contextual information, obtaining global features with stronger feature representation. The Parallel Attention Fusion Module employs an attention mechanism to fuse global and local features in both channel and spatial dimensions in parallel, enriching the semantic relations extracted during unsupervised training and improving the performance of unsupervised semantic segmentation. K-means clustering is then performed on the fused features to achieve high-precision unsupervised semantic segmentation. Experiments with MSHFE-Net on the Potsdam and Vaihingen datasets demonstrate its effectiveness in significantly improving the accuracy of unsupervised semantic segmentation.
Förderung
- Natural Science Foundation of China
- Outstanding Youth Science Fund of Xi’an University of Science and Technology
- Scientific Research Plan Projects of Shaanxi Education Department
- China Postdoctoral Science Foundation
- Natural Science Basic Research Plan in Shaanxi Province of China
- Civil Space Thirteen Five Years Pre-Research Project
Zitieren
APA
Song, W., Nie, F., Wang, C., Jiang, Y., & Wu, Y. (2024). Unsupervised Multi-Scale Hybrid Feature Extraction Network for Semantic Segmentation of High-Resolution Remote Sensing Images. Remote Sensing, 16(20), 3774. https://doi.org/10.3390/rs16203774
Vancouver
Song W, Nie F, Wang C, Jiang Y, Wu Y. Unsupervised Multi-Scale Hybrid Feature Extraction Network for Semantic Segmentation of High-Resolution Remote Sensing Images. Remote Sensing. 2024;16(20):3774. doi:10.3390/rs16203774
BibTeX
@article{Song2024,
author = {Song, Wanying and Nie, Fangxin and Wang, Chi and Jiang, Yinyin and Wu, Yan},
title = {Unsupervised Multi-Scale Hybrid Feature Extraction Network for Semantic Segmentation of High-Resolution Remote Sensing Images},
journal = {Remote Sensing},
year = {2024},
volume = {16},
number = {20},
pages = {3774},
doi = {10.3390/rs16203774}
}
EUVIMED