Urban flood mapping using Sentinel-1 and RADARSAT Constellation Mission image and convolutional Siamese network

Urban floods can affect people's lives and properties, and therefore, urban flood mapping is crucial for reliable risk assessment and the development of effective mitigation strategies. With the advent of high spatial and temporal resolution satellite images, remote sensing has become popular for urban flood mapping. Synthetic aperture RADAR (SAR) sensors can capture image data during a flood event because their emitted signal can penetrate through the clouds. However, they have some limitations, such as layover, shadowing, and speckle noise, that might challenge their usage, especially for urban flood mapping. Deep learning (DL) algorithms have been widely used for automatic urban flood mapping using remote sensing data, but the flood mapping accuracy achieved using SAR and DL algorithms is still uncertain. This paper proposes a DL-based change detection framework, convolutional Siamese network (CSN), for flood mapping in three urban areas: parts of Ottawa, ON and Gatineau, QC, Abbotsford, BC, and Leverkusen, Germany. The datasets applied were Sentinel-1 and dual-polarized RADARSAT Constellation Mission (RCM) data. The applied data were captured in C-band, and their resolutions were 10 m and 5 m for Sentinel-1 and RCM, respectively. Comparison with other DL-based segmentation algorithms, including Unet, Unet++, DeepLabV3+, and Siamese-Unet, confirmed the reliability of the proposed CSN. Although a promising flood recall rate of about 0.7 was achieved, it was inferred from the flood precision and F1 score that Sentinel-1 data medium resolution might hinder its application for urban flood mapping. Further, RCM data were also tested in both urban and non-urban areas, and a precision of 0.79 was achieved for the non-urban case. Experiments on two existing datasets, SEN12FLOOD and SEN1FLOOD11, showed that the proposed CSN achieved a higher precision index of 0.75 on SEN12-FLOOD than SEN1FLOOD11 dataset, with a precision of 0.2, because of different labeling formats between the two datasets.