TSNet: A Two-stage Segmentation Network Guided by Uncertainty for Remote Sensing Image Road Extraction

Jiajia Liu; Xuan Zhao; Wenxiang Dong; Jianhua Liu; Xiaoguang Tu

doi:10.54097/nvegp737

Authors

Jiajia Liu Civil Aviation Flight University of China, Center for Faculty Development and Teaching Evaluation, Guanghan 618307, China
Xuan Zhao Civil Aviation Flight University of China, College of Aviation Electronic and Electrical Engineering, Guanghan 618307, China
Wenxiang Dong Civil Aviation Flight University of China, College of Aviation Electronic and Electrical Engineering, Guanghan 618307, China
Jianhua Liu Civil Aviation Flight University of China, College of Aviation Electronic and Electrical Engineering, Guanghan 618307, China
Xiaoguang Tu Civil Aviation Flight University of China, College of Aviation Electronic and Electrical Engineering, Guanghan 618307, China

DOI:

https://doi.org/10.54097/nvegp737

Keywords:

Remote sensing imagery; Road extraction; Semantic segmentation; Two-stage network; Uncertainty-aware.

Abstract

Road extraction from high‑resolution remote‑sensing images plays a vital practical role in multiple application scenarios including urban layout planning and autonomous driving systems. Nevertheless, current road‑extraction approaches still suffer from several bottlenecks under complex ground environments, including indistinct object boundaries, missing fine‑grained features and fragmented road outputs. Aiming to mitigate the above defects, this paper puts forward an uncertainty‑guided two‑stage segmentation network termed TSNet. The proposed framework divides road segmentation into two successive subtasks: global‑level semantic understanding and local boundary optimization. A cross‑layer residual feature enhancement module (CRM) is adopted to extract and aggregate multi‑scale feature information. On the basis of multi‑scale fused features, the first segmentation stage produces preliminary probability maps, and an uncertainty‑aware learning module (UALM) is introduced to measure the confidence degree for each pixel‑wise prediction. In the subsequent second stage, the local refinement module (LRM) takes the generated uncertainty map as guidance to dynamically concentrate on regions prone to prediction errors, and implements adaptive optimization for road edge areas. Quantitative experiments on the DeepGlobe dataset prove that the presented model achieves competitive performance in maintaining road connectivity, yielding Accuracy, Precision, Recall, F1‑score and Intersection over Union (IoU) of 98.68%, 83.72%, 82.36%, 82.85% and 71.07%, respectively. Further validation on the CHN6‑CUG dataset obtains an Accuracy of 96.10%, Precision of 80.48%, Recall of 79.56%, F1‑score of 79.37% and IoU of 66.20%. When compared with state‑of‑the‑art competing algorithms, our method achieves optimal results for all five evaluation metrics. In addition, qualitative visualization results reveal that the model exhibits prominent strengths in retaining connected road topology and recovering intricate edge contours, which effectively enhances the robustness and prediction accuracy of road extraction for complex real‑world scenes.

Downloads

Download data is not yet available.

References

[1] Liu, R., Wu, J., Lu, W., Miao, Q., Zhang, H., Liu, X., Lu, Z., & Li, L. (2024). A review of deep learning-based methods for road extraction from high-resolution remote sensing images. Remote Sensing, 16(12), 2056. https://doi.org/10.3390/rs16122056. DOI: https://doi.org/10.3390/rs16122056

[2] Abdollahi, A., Pradhan, B., Shukla, N., Chakraborty, S., & Alamri, A. (2022). Deep learning approaches applied to remote sensing datasets for road extraction: A state-of-the-art review. IEEE Access, 10, 45205–45239. https://doi.org/10.1109/ACCESS.2022.3169872

[3] Ronneberger, O., Fischer, P., & Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015 (pp. 234–241). Springer International Publishing. https://doi.org/10.1007/978-3-319-24574-4_28 DOI: https://doi.org/10.1007/978-3-319-24574-4_28

[4] Yang, X., Zhang, Y. M., Wang, L., & Zhao, J. H. (2022). Road extraction from high-resolution remote sensing images based on improved U-Net. Journal of Geomatics, 47(3), 1–6.

[5] Chen, L. C., Zhu, Y., Papandreou, G., Schroff, F., & Adam, H. (2018). Encoder-decoder with atrous separable convolution for semantic image segmentation. In Proceedings of the European Conference on Computer Vision (ECCV) (pp. 801–818). https://doi.org/10.1007/978-3-030-01234-2_49 DOI: https://doi.org/10.1007/978-3-030-01234-2_49

[6] Zhou, L., Zhang, C., & Wu, M. (2018). D-LinkNet: LinkNet with pretrained encoder and dilated convolution for high resolution satellite imagery road extraction. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops (pp. 182–186). DOI: https://doi.org/10.1109/CVPRW.2018.00034

[7] Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., & Guo, B. (2021). Swin transformer: Hierarchical vision transformer using shifted windows. In Proceedings of the IEEE/CVF International Conference on Computer Vision (pp. 10012–10022). https://doi.org/10.1109/ICCV48922.2021.00986 DOI: https://doi.org/10.1109/ICCV48922.2021.00986

[8] Xie, E., Wang, W., Yu, Z., Anandkumar, M., Alvarez, J. M., & Luo, P. (2021). SegFormer: Simple and efficient design for semantic segmentation with transformers. Advances in Neural Information Processing Systems, 34, 12077–12090.

[9] Tao, J., Chen, Z., Sun, Z. C., Guo, H. D., Zeng, B. L., Yu, Z. B., Wang, Y. L., He, Z. Q., & Lei, X. Q. (2023). SEG-Road: A segmentation network for road extraction based on transformer and CNN with connectivity structures. Remote Sensing, 15(6), 1602. https://doi.org/10.3390/rs15061602 DOI: https://doi.org/10.3390/rs15061602

[10] Guo, Z., Bian, L., Wei, H., Li, J., Ni, H., & Huang, X. (2024). DSNet: A novel way to use atrous convolutions in semantic segmentation. IEEE Transactions on Circuits and Systems for Video Technology, 35, 3679–3692. https://doi.org/10.1109/TCSVT.2024.3401821 DOI: https://doi.org/10.1109/TCSVT.2024.3509504

[11] Yang, Z., Bi, F., Hou, X., Zhou, D., & Wang, Y. (2024). DDRNet: Dual-domain refinement network for remote sensing image semantic segmentation. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 17, 20177–20189. https://doi.org/10.1109/JSTARS.2024.3422347 DOI: https://doi.org/10.1109/JSTARS.2024.3490584

[12] Liu, J., & Huang, C. (2024). Multi-scale adaptive road extraction network for high-resolution remote sensing images. Frontiers in Computational Intelligence and Systems, 10(3), 101–106. DOI: https://doi.org/10.54097/cp052n67

[13] He, F., Liu, S., Liu, S., Jin, Y., Xie, H., & Tong, X. (2025). HDRoad: An encoder-decoder architecture with hybrid attention and directional prior for efficient road extraction from remote sensing images. ISPRS Journal of Photogrammetry and Remote Sensing, 227, 251–264. https://doi.org/10.1016/j.isprsjprs.2025.06.014 DOI: https://doi.org/10.1016/j.isprsjprs.2025.06.014

[14] Ruby, U., & Yendapalli, V. (2020). Binary cross entropy with deep learning technique for image classification. International Journal of Advanced Trends in Computer Science and Engineering, 9(10). DOI: https://doi.org/10.30534/ijatcse/2020/175942020

[15] Sudre, C. H., Li, W., Vercauteren, T., Ourselin, S., & Cardoso, M. J. (2017). Generalised dice overlap as a deep learning loss function for highly unbalanced segmentations. In Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support (pp. 240–248). Springer International Publishing. https://doi.org/10.1007/978-3-319-67558-9_28 DOI: https://doi.org/10.1007/978-3-319-67558-9_28

[16] Yu, Y., Wang, C., Fu, Q., Kou, R., Wu, W., & Liu, T. (2023). Survey of evaluation metrics and methods for semantic segmentation. Computer Engineering and Applications, 59(6), 57–69.

[17] Li, R., Wang, L., Zhang, C., Duan, C., & Zheng, S. (2022). A2-FPN for semantic segmentation of fine-resolution remotely sensed images. International Journal of Remote Sensing, 43(3), 1131–1155. https://doi.org/10.1080/01431161.2021.2022775 DOI: https://doi.org/10.1080/01431161.2022.2030071

[18] Bo, W., Liu, J., Fan, X., Tjahjadi, T., Ye, Q., & Fu, L. (2022). BASNet: Burned area segmentation network for real-time detection of damage maps in remote sensing images. IEEE Transactions on Geoscience and Remote Sensing, 60, 1–13. https://doi.org/10.1109/TGRS.2022.3141203 DOI: https://doi.org/10.1109/TGRS.2022.3197647

[19] Mehta, D., Skliar, A., Ben Yahia, H., Borse, S., Porikli, F., Habibian, A., & Blankevoort, T. (2022). Simple and efficient architectures for semantic segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 2628–2636). IEEE. DOI: https://doi.org/10.1109/CVPRW56347.2022.00296

[20] Huang, H., Lin, L., Tong, R., Hu, H., Zhang, Q., & Iwamoto, Y. (2020). Unet 3+: A full-scale connected unet for medical image segmentation. In ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 1055–1059). IEEE. https://doi.org/10.1109/ICASSP40776.2020.9053405 DOI: https://doi.org/10.1109/ICASSP40776.2020.9053405

[21] Wang, L., Li, R., Zhang, C., Fang, S., Duan, C., Meng, X., & Atkinson, P. M. (2022). UNetFormer: A UNet-like transformer for efficient semantic segmentation of remote sensing urban scene imagery. In ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 190–214). IEEE. https://doi.org/10.1109/ICASSP43922.2022.9746115 DOI: https://doi.org/10.1016/j.isprsjprs.2022.06.008

[22] Yuan, F., Wang, Z. Y., & Lu, K. (2025). DeepLabv3+ remote sensing image road extraction based on multilevel feature fusion. Laser Journal, 46(7), 93–100.