CSNet: A Remote Sensing Image Semantic Segmentation Network Based on Coordinate Attention and Skip Connections
Remote Sensing · 13.06.2025 · 17 (12) · 2048
DOI: 10.3390/rs17122048 ↗Kurz gefasst
Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 13.06.2025. Verfasst von Jiahao Li und 4 weiteren Autor:innen. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Sichuan Science and Technology Program “Research on Precise Identification of Forest Fire Risks for Distribution Lines of 35 kV and Below Crossing Forest Areas”.
Abstract (Originalsprache)
In recent years, the continuous development of deep learning has significantly advanced its application in the field of remote sensing. However, the semantic segmentation of high-resolution remote sensing images remains challenging due to the presence of multi-scale objects and intricate spatial details, often leading to the loss of critical information during segmentation. To address this issue and enable fast and accurate segmentation of remote sensing images, we made improvements based on SegNet and named the enhanced model CSNet. CSNet is built upon the SegNet architecture and incorporates a coordinate attention (CA) mechanism, which enables the network to focus on salient features and capture global spatial information, thereby improving segmentation accuracy and facilitating the recovery of spatial structures. Furthermore, skip connections are introduced between the encoder and decoder to directly transfer low-level features to the decoder. This promotes the fusion of semantic information at different levels, enhances the recovery of fine-grained details, and optimizes the gradient flow during training, effectively mitigating the vanishing gradient problem and improving training efficiency. Additionally, a hybrid loss function combining weighted cross-entropy and Dice loss is employed. To address the issue of class imbalance, several categories within the dataset are merged, and samples with an excessively high proportion of background pixels are removed. These strategies significantly enhance the segmentation performance, particularly for small-sample classes. Experimental results from the Five-Billion-Pixels dataset demonstrate that, while introducing only a modest increase in parameters compared to SegNet, CSNet achieves superior segmentation performance in terms of overall classification accuracy, boundary delineation, and detail preservation, outperforming established methods such as U-Net, FCN, DeepLabv3+, SegNet, ViT, HRNe and BiFormert.
Förderung
- Sichuan Science and Technology Program “Research on Precise Identification of Forest Fire Risks for Distribution Lines of 35 kV and Below Crossing Forest Areas”
Zitieren
APA
Li, J., Zhang, H., Chen, L., He, B., & Chen, H. (2025). CSNet: A Remote Sensing Image Semantic Segmentation Network Based on Coordinate Attention and Skip Connections. Remote Sensing, 17(12), 2048. https://doi.org/10.3390/rs17122048
Vancouver
Li J, Zhang H, Chen L, He B, Chen H. CSNet: A Remote Sensing Image Semantic Segmentation Network Based on Coordinate Attention and Skip Connections. Remote Sensing. 2025;17(12):2048. doi:10.3390/rs17122048
BibTeX
@article{Li2025,
author = {Li, Jiahao and Zhang, Hongguo and Chen, Liang and He, Binbin and Chen, Huaixin},
title = {CSNet: A Remote Sensing Image Semantic Segmentation Network Based on Coordinate Attention and Skip Connections},
journal = {Remote Sensing},
year = {2025},
volume = {17},
number = {12},
pages = {2048},
doi = {10.3390/rs17122048}
}
EUVIMED