Boundary-Aware Semantic Segmentation of Remote Sensing Images via Segformer and Snake Convolution
- Geely University of China
Abstract
Semantic segmentation of remote sensing images remains challenging due to complex object structures and varying scales. This paper proposes a novel hybrid segmentation model that combines Segformer for global context extraction with Dynamic Snake Convolution to better capture fine-grained, boundary-aware features. An auxiliary semantic branch is introduced to improve feature alignment across scales. Experiments on three benchmark datasets—LoveDA, Potsdam, and Vaihingen—demonstrate that the proposed approach achieves consistent improvements in mIoU over baseline models, particularly in segmenting irregular and linear structures. This framework offers a promising solution for high-resolution land cover mapping and urban scene understanding.
Key words
Segformer, dynamic snake convolution, remote sensing, Semantic Segmentation, Deep learning
Digital Object Identifier (DOI)
https://doi.org/10.2298/CSIS250312054X
Publication information
Volume 22, Issue 3 (June 2025)
Year of Publication: 2025
ISSN: 2406-1018 (Online)
Publisher: ComSIS Consortium
Full text
Available in PDF
Portable Document Format
How to cite
Xia, Y., Zhang, L., Guo, T., Qi, J.: Boundary-Aware Semantic Segmentation of Remote Sensing Images via Segformer and Snake Convolution. Computer Science and Information Systems, Vol. 22, No. 3, 991–1010. (2025), https://doi.org/10.2298/CSIS250312054X