Urban-aware U-Net for large-scale urban flood mapping using multitemporal Sentinel-1 intensity and interferometric coherence

Zhao, J.; Li, Y.; Matgen, P.; Pelich, R.; /Hostache, Renaud; Wagner, W.; Chini, M.

DSpace Principal
→
Colección de los Centros Asociados
→
Francia
→
Institut de Recherche pour le Développement - IRD - Cosecha
→
Ver ítem

dc.creator	Zhao, J.
dc.creator	Li, Y.
dc.creator	Matgen, P.
dc.creator	Pelich, R.
dc.creator	/Hostache, Renaud
dc.creator	Wagner, W.
dc.creator	Chini, M.
dc.date	2022
dc.date.accessioned	2022-10-14T18:38:23Z
dc.date.available	2022-10-14T18:38:23Z
dc.identifier	https://www.documentation.ird.fr/hor/fdi:010086053
dc.identifier	oai:ird.fr:fdi:010086053
dc.identifier	Zhao J., Li Y., Matgen P., Pelich R., Hostache Renaud, Wagner W., Chini M.. Urban-aware U-Net for large-scale urban flood mapping using multitemporal Sentinel-1 intensity and interferometric coherence. 2022, 60, p. 4209121 [21 p.]
dc.identifier.uri	https://biblioteca-repositorio.clacso.edu.ar/handle/CLACSO/169877
dc.description	Due to the complexity of backscattering mechanisms in built-up areas, the synthetic aperture radar (SAR)-based mapping of floodwater in urban areas remains challenging. Open areas affected by flooding have low backscatter due to the specular reflection of calm water surfaces. Floodwater within built-up areas leads to double-bounce effects, the complexity of which depends on the configuration of floodwater concerning the facades of the surrounding buildings. Hence, it has been shown that the analysis of interferometric SAR coherence reduces the underdetection of floods in urbanized areas. Moreover, the high potential of deep convolutional neural networks for advancing SAR-based flood mapping is widely acknowledged. Therefore, we introduce an urban-aware U-Net model using dual-polarization Sentinel-1 multitemporal intensity and coherence data to map the extent of flooding in urban environments. It uses a priori information (i.e., an SAR-derived probabilistic urban mask) in the proposed urban-aware module, consisting of channel-wise attention and urban-aware normalization submodules to calibrate features and improve the final predictions. In this study, Sentinel-1 single-look complex data acquired over four study sites from three continents have been considered. The qualitative evaluation and quantitative analysis have been carried out using six urban flood cases. A comparison with previous methods reveals a significant enhancement in the accuracy of urban flood mapping: the F1 score of flooded urban increased from 0.3 to 0.6 with few false alarms in urban area using our method. Experimental results indicate that the proposed model trained with limited datasets has strong potential for near-real-time urban flood mapping.
dc.language	EN
dc.subject	Floods
dc.subject	Coherence
dc.subject	Synthetic aperture radar
dc.subject	Urban areas
dc.subject	Buildings
dc.subject	Training
dc.subject	Soil
dc.subject	Deep learning (DL)
dc.subject	multitemporal synthetic aperture radar (SAR)
dc.subject	U-Net
dc.subject	urban flood mapping
dc.subject	urban-aware
dc.title	Urban-aware U-Net for large-scale urban flood mapping using multitemporal Sentinel-1 intensity and interferometric coherence
dc.type	text
dc.coverage	ETATS UNIS
dc.coverage	JAPON
dc.coverage	SOMALIE
dc.coverage	MOZAMBIQUE