TMSA-Net: Transformer-Based Multi-Scale Attention U-Net for Flood Image Segmentation

Parham Imanzadeh Charandabi , Shivakumara Palaiahnakote , Apostolos Antonacopoulos , Christian Clausner

CAAI Transactions on Intelligence Technology ›› 2026, Vol. 11 ›› Issue (4) : 1104 -1126.

PDF (8206KB)
CAAI Transactions on Intelligence Technology ›› 2026, Vol. 11 ›› Issue (4) :1104 -1126. DOI: 10.1049/cit2.70169
ORIGINAL RESEARCH
research-article
TMSA-Net: Transformer-Based Multi-Scale Attention U-Net for Flood Image Segmentation
Author information +
History +
PDF (8206KB)

Abstract

Flood detection is essential for real-time applications, including disaster management, emergency response, and alerting people in flood zones. For successful flood detection, accurate flood region segmentation is essential. However, the flood region segmentation is challenging due to the complex background and occlusions with debris and the effect of external adverse factors. In contrast to existing models that focus on satellite and remote sensing images, for real-time applications, this study proposes a new Transformer-based Multi-Scale-Attention U-Net (TMSA-Net) model for flood region segmentation in images with cluttered backgrounds. The proposed model extracts features hierarchically using a Convolutional Neural Network (CNN) encoder. The deepest features are fed to an adapted vision transformer for extracting global context. In the decoder, Multi-Scale Attention Gates are used in skip connections to selectively filter encoder features before fusion. The proposed model is compared with state-of-the-art methods, including Swin-UNet, DeepLabV3+, TransUNet, UNet++, and Attention U-Net. Experimental results show that TMSA-Net achieves a test IoU of 92.01%, with improvements of 2.82% and 5.95% IoU over DeepLabV3+ and TransUNet, respectively, demonstrating the effectiveness of the proposed approach. The newly created experimental dataset, with ground truth at the pixel level, will be publicly available (released upon acceptance).

Keywords

attention mechanism / deep learning / disaster response / flood detection / multi-scale processing / semantic segmentation / vision transformer

Cite this article

Download citation ▾
Parham Imanzadeh Charandabi, Shivakumara Palaiahnakote, Apostolos Antonacopoulos, Christian Clausner. TMSA-Net: Transformer-Based Multi-Scale Attention U-Net for Flood Image Segmentation. CAAI Transactions on Intelligence Technology, 2026, 11 (4) : 1104-1126 DOI:10.1049/cit2.70169

登录浏览全文

4963

注册一个新账户 忘记密码

References

[1]

A. Domeneghetti, “ On the Use of SRTM and Altimetry Data for Flood Modeling in Data-Sparse Regions,” Water Resources Research 52, no. 4 (2016): 2901-2918, https://doi.org/10.1002/2015WR017967.

[2]

Y. Hirabayashi, R. Mahendran, S. Koirala, et al., “ Global Flood Risk Under Climate Change,” Nature Climate Change 3, no. 9 (2013): 816-821, https://doi.org/10.1038/nclimate1911.

[3]

D. Bonafilia, B. Tellman, T. Anderson, and E. Issenberg, “ Sen1Floods11: A Georeferenced Dataset to Train and Test Deep Learning Flood Algorithms for Sentinel-1,” in 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (IEEE Xplore, 2020), 835-845, https://doi.org/10.1109/CVPRW50498.2020.00113.

[4]

M. S. Rahman and L. Di, “ The State of the Art of Spaceborne Remote Sensing in Flood Management,” Natural Hazards 85, no. 2 (2017): 1223-1248, https://doi.org/10.1007/s11069-016-2601-9.

[5]

N. Tsutsumida, T. Tanaka, and N. Sultana, “ Automated Flood Detection From Sentinel-1 GRD Time Series Using Bayesian Analysis for Change Point Problems,” IEEE Transactions on Geoscience and Remote Sensing 63 (2025): 1-13, https://doi.org/10.1109/TGRS.2025.3620907.

[6]

M. Erdelj, E. Natalizio, K. R. Chowdhury, and I. F. Akyildiz, “ Help From the Sky: Leveraging UAVs for Disaster Management,” IEEE Pervasive Computing 16, no. 1 (2017): 24-32, https://doi.org/10.1109/MPRV.2017.11.

[7]

K. Amini, Y. Liu, J. E. Padgett, G. Balakrishnan, and A. Veeraraghavan, “ Debris Segmentation Using Post-Hurricane Aerial Imagery,” Computer-Aided Civil and Infrastructure Engineering 40, no. 25 (2025): 4116-4131, https://doi.org/10.1111/mice.70033.

[8]

Y. LeCun, Y. Bengio, and G. Hinton, “ Deep Learning,” Nature 521, no. 7553 (2015): 436-444, https://doi.org/10.1038/nature14539.

[9]

O. Ronneberger, P. Fischer, and T. Brox, “ U-Net: Convolutional Networks for Biomedical Image Segmentation,” arXiv (2015), https://doi.org/10.48550/arXiv.1505.04597.

[10]

L. Zhou, T. Chen, F. Yang, et al., “ MSFlood-Net: A Physically Informed Deep Learning Model Integrating Multi-Source Data for Flood Inundation Mapping,” Environmental Modelling & Software 196 (2026): 106779, https://doi.org/10.1016/j.envsoft.2025.106779.

[11]

O. Oktay, J. Schlemper, L. L. Folgoc, et al., “ Attention U-Net: Learning Where to Look for the Pancreas,” arXiv (2018), https://doi.org/10.48550/arXiv.1804.03999.

[12]

S. Woo, J. Park, J.-Y. Lee, and I. S. Kweon, “ CBAM: Convolutional Block Attention Module,” arXiv (2018), https://doi.org/10.48550/arXiv.1807.06521.

[13]

F. Fakhri and I. Gkanatsios, “ Quantitative Evaluation of Flood Extent Detection Using Attention U-Net Case Studies From Eastern South Wales Australia in March 2021 and July 2022,” Scientific Reports 15, no. 1 (2025): 12377, https://doi.org/10.1038/s41598-025-92734-x.

[14]

A. Refice, D. Capolongo, M. Chini, and A. D’Addabbo, “ Improving Flood Detection and Monitoring Through Remote Sensing,” Water 14, no. 3 (2022): 364, https://doi.org/10.3390/w14030364.

[15]

E. Nemni, J. Bullock, S. Belabbes, and L. Bromley, “ Fully Convolutional Neural Network for Rapid Flood Segmentation in Synthetic Aperture Radar Imagery,” Remote Sensing 12, no. 16 (2020): 2532, https://doi.org/10.3390/rs12162532.

[16]

M. Siddique and T. Ahmed, “ CCD-Conv1D: A Deep Learning Based Coherent Change Detection Technique to Monitor and Forecast Floods Using Sentinel-1 Images,” Remote Sensing Applications: Society and Environment 37 (2025): 101440, https://doi.org/10.1016/j.rsase.2024.101440.

[17]

A. Krizhevsky, I. Sutskever, and G. E. Hinton, “ ImageNet Classification With Deep Convolutional Neural Networks,” Communications of the ACM 60, no. 6 (2017): 84-90, https://doi.org/10.1145/3065386.

[18]

K. Simonyan and A. Zisserman, “ Very Deep Convolutional Networks for Large-Scale Image Recognition,” arXiv (2015), https://doi.org/10.48550/arXiv.1409.1556.

[19]

J. Long, E. Shelhamer, and T. Darrell, “ Fully Convolutional Networks for Semantic Segmentation,” arXiv (2015), https://doi.org/10.48550/arXiv.1411.4038.

[20]

E. Shelhamer, J. Long, and T. Darrell, “ Fully Convolutional Networks for Semantic Segmentation,” IEEE Transactions on Pattern Analysis and Machine Intelligence 39, no. 4 (2017): 640-651, https://doi.org/10.1109/TPAMI.2016.2572683.

[21]

V. Badrinarayanan, A. Kendall, and R. Cipolla, “ SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation,” arXiv (2016), https://doi.org/10.48550/arXiv.1511.00561.

[22]

H. Zhao, J. Shi, X. Qi, X. Wang, and J. Jia, “ Pyramid Scene Parsing Network,” in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (IEEE Xplore, 2017), 6230-6239, https://doi.org/10.1109/CVPR.2017.660.

[23]

B. Bahrami and H. Arbabkhah, “ Enhanced Flood Detection Through Precise Water Segmentation Using Advanced Deep Learning Models,” Journal of Civil Engineering Research 6, no. 1 (2024): 1-8, https://doi.org/10.61186/JCER.6.1.1.

[24]

F. I. Diakogiannis, F. Waldner, P. Caccetta, and C. Wu, “ ResUNet-a: A Deep Learning Framework for Semantic Segmentation of Remotely Sensed Data,” ISPRS Journal of Photogrammetry and Remote Sensing 162 (2020): 94-114, https://doi.org/10.1016/j.isprsjprs.2020.01.013.

[25]

Zhou, Z., M. M. R. Siddiquee, N. Tajbakhsh, and J. Liang, “ UNet++: A Nested U-Net Architecture for Medical Image Segmentation,” arXiv (2018), https://doi.org/10.48550/arXiv.1807.10165

[26]

H. Huang, L. Lin, R. Tong, et al., “ 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), (2020), 1055-1059, https://doi.org/10.1109/ICASSP40776.2020.9053405.

[27]

G. Zhao, Y. Zhang, M. Ge, and M. Yu, “ Bilateral U-Net Semantic Segmentation With Spatial Attention Mechanism,” CAAI Transactions on Intelligence Technology 8, no. 2 (2023): 297-307, https://doi.org/10.1049/cit2.12118.

[28]

N. S. Jonnala, R. C. Bheemana, K. Prakash, et al., “ DSIA U-Net: Deep Shallow Interaction With Attention Mechanism UNet for Remote Sensing Satellite Images,” Scientific Reports 15, no. 1 (2025): 549, https://doi.org/10.1038/s41598-024-84134-4.

[29]

N. S. Jonnala, S. Siraaj, Y. Prastuti, et al., “ AER U-Net: Attention-Enhanced Multi-Scale Residual U-Net Structure for Water Body Segmentation Using Sentinel-2 Satellite Images,” Scientific Reports 15, no. 1 (2025): 16099, https://doi.org/10.1038/s41598-025-99322-z.

[30]

L.-C. Chen, G. Papandreou, F. Schroff, and H. Adam, “ Rethinking Atrous Convolution for Semantic Image Segmentation,” arXiv (2017), https://doi.org/10.48550/arXiv.1706.05587.

[31]

A. A. Sundaresan and A. A. Solomon, “ Post-Disaster Flooded Region Segmentation Using DeepLabv3+ and Unmanned Aerial System Imagery,” Natural Hazards Research 5, no. 2 (2025): 363-371, https://doi.org/10.1016/j.nhres.2024.12.003.

[32]

M. U. Danish, M. Buwaneswaran, T. Fonseka, and K. Grolinger, “ Graph Attention Convolutional U-NET: A Semantic Segmentation Model for Identifying Flooded Areas,” arXiv (2025), https://doi.org/10.48550/arXiv.2502.15907.

[33]

N. K. Sharma and M. Saharia, “ DeepSARFlood: Rapid and Automated SAR-Based Flood Inundation Mapping Using Vision Transformer-Based Deep Ensembles With Uncertainty Estimates,” Science of Remote Sensing 11 (2025): 100203, https://doi.org/10.1016/j.srs.2025.100203.

[34]

F. Alam, F. Ofli, and M. Imran, “ CrisisMMD: Multimodal Twitter Datasets From Natural Disasters,” in Proceedings of the International AAAI Conference on Web and Social Media 12, no. 1 (2018), https://doi.org/10.1609/icwsm.v12i1.14983.

[35]

A. Hussain, G. Latif, J. Alghazo, and E. Kim, “ Flood Detection Using Deep Learning Methods From Visual Images,” in AIP Conference Proceedings 3063 (2024): 030004, https://doi.org/10.1063/5.0194669.

[36]

M. Imran, C. Castillo, F. Diaz, and S. Vieweg, “ Processing Social Media Messages in Mass Emergency: Survey Summary,” in Companion of the the Web Conference 2018 on the Web Conference 2018-WWW’18, (2018), 507-511, https://doi.org/10.1145/3184558.3186242.

[37]

Y. Wang, Y. Shen, B. Salahshour, et al., “ Urban Flood Extent Segmentation and Evaluation From Real-World Surveillance Camera Images Using Deep Convolutional Neural Network,” Environmental Modelling & Software 173 (2024): 105939, https://doi.org/10.1016/j.envsoft.2023.105939.

[38]

A. Vaswani, N. Shazeer, N. Parmar, et al., “ Attention Is All You Need,” arXiv (2023), https://doi.org/10.48550/arXiv.1706.03762.

[39]

A. Dosovitskiy, L. Beyer, A. Kolesnikov, et al., “ An Image Is Worth 16x16 Words: Transformers for Image Recognition at Scale,” arXiv (2021), https://doi.org/10.48550/arXiv.2010.11929.

[40]

J. Chen, Y. Lu, Q. Yu, et al., “ TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation,” arXiv (2021), https://doi.org/10.48550/arXiv.2102.04306.

[41]

H. Cao, Y. Wang, J. Chen, et al., “ Swin-Unet: Unet-Like Pure Transformer for Medical Image Segmentation,” arXiv (2021), https://doi.org/10.48550/arXiv.2105.05537.

[42]

Y. Gao, M. Zhou, and D. Metaxas, “ UTNet: A Hybrid Transformer Architecture for Medical Image Segmentation,” arXiv (2021), https://doi.org/10.48550/arXiv.2107.00781.

[43]

Y. Zhang, H. Liu, and Q. Hu, “ TransFuse: Fusing Transformers and CNNs for Medical Image Segmentation,” arXiv (2021), https://doi.org/10.48550/arXiv.2102.08005.

[44]

Y. Xie, J. Zhang, C. Shen, and Y. Xia, “ CoTr: Efficiently Bridging CNN and Transformer for 3D Medical Image Segmentation,” arXiv (2021), https://doi.org/10.48550/arXiv.2103.03024.

[45]

B. Yu, Q. Zhou, L. Yuan, H. Liang, P. Shcherbakov, and X. Zhang, “ 3D Medical Image Segmentation Using the Serial-Parallel Convolutional Neural Network and Transformer Based on Cross-Window Self-Attention,” CAAI Transactions on Intelligence Technology 10, no. 2 (2025): 337-348, https://doi.org/10.1049/cit2.12411.

[46]

J. Hu, L. Shen, S. Albanie, G. Sun, and E. Wu, “ Squeeze-And-Excitation Networks,” arXiv (2019), https://doi.org/10.48550/arXiv.1709.01507.

[47]

Y. Zhang, Q. Yang, T. Li, et al., “ V-UNet: Medical Image Segmentation Based on Variational Attention Mechanism,” CAAI Transactions on Intelligence Technology 10, no. 5 (2025): 1350-1362, https://doi.org/10.1049/cit2.70053.

[48]

M. Tan, and Q. V. Le, “ EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks,” in Proceedings of the 36th International Conference on Machine Learning, Vol. 97 (PMLR, 2019).

[49]

M. Rahnemoonfar, T. Chowdhury, A. Sarkar, D. Varshney, M. Yari, and R. R. Murphy, “ FloodNet: A High Resolution Aerial Imagery Dataset for Post Flood Scene Understanding,” IEEE Access 9 (2021): 89644-89654, https://doi.org/10.1109/ACCESS.2021.3090981.

[50]

M. Rahnemoonfar, T. Chowdhury, and R. Murphy, “ RescueNet: A High Resolution UAV Semantic Segmentation Dataset for Natural Disaster Damage Assessment,” Scientific Data 10, no. 1 (2023): 913, https://doi.org/10.1038/s41597-023-02799-4.

[51]

R. Karanjit, R. Pally, and V. Samadi, “ Flooding Images Dataset,” (2023) [Dataset], https://doi.org/10.34740/kaggle/ds/1864149.

[52]

L. Yang, “ Flood Semantic Segmentation Dataset,” Kaggle, (2024) [Dataset], https://www.kaggle.com/datasets/lihuayang111265/flood-semantic-segmentation-dataset.

PDF (8206KB)

0

Accesses

0

Citation

Detail

Sections
Recommended

/

〈 〉