Sentinel-1-Based Water and Flood Mapping: Benchmarking Convolutional Neural Networks Against an Operational Rule-Based Processing Chain

Abstract

In this study, the effectiveness of several convolutional neural network architectures (AlbuNet-34/FCN/DeepLabV3+/U-Net/U-Net++) for water and flood mapping using Sentinel-1 amplitude data is compared to an operational rule-based processor (S-1FS). This comparison is made using a globally distributed dataset of Sentinel-1 scenes and the corresponding ground truth water masks derived from Sentinel-2 data to evaluate the performance of the classifiers on a global scale in various environmental conditions. The impact of using single versus dual-polarized input data on the segmentation capabilities of AlbuNet-34 is evaluated. The weighted cross entropy loss is combined with the Lovász loss and various data augmentation methods are investigated. Furthermore, the concept of atrous spatial pyramid pooling used in DeepLabV3+ and the multi-scale feature fusion inherent in U-Net++ are assessed. Finally, the generalization capacity of AlbuNet-34 is tested in a realistic flood mapping scenario by using additional data from two flood events and the Sen1Floods11 dataset. The model trained using dual polarized data outperforms the S-1FS significantly and increases the intersection over union (IoU) score by 5 %. Using a weighted combination of the cross entropy and the Lovász loss increases the IoU score by another 2 %. Geometric data augmentation degrades the performance while radiometric data augmentation leads to better testing results. FCN/DeepLabV3+/U-Net/U-Net++ perform not significantly different to AlbuNet-34. Models trained on data showing no distinct inundation perform very well in mapping the water extent during two flood events, reaching IoU scores of 0.96 and 0.94 respectively, and perform comparatively well on the Sen1Floods11 dataset.

Cross-modal distillation for flood extent mapping

“…In Table 4, Sen1Floods11 weak-label baseline uses FCNN (Long et al, 2015) architecture with ResNet 50 (He et al, 2016) encoder backbone. Similarly, AN-34 (Helleis et al, 2022) and S-1FS (Helleis et al, 2022) use UNet-based (Ronneberger et al, 2015) encoder–decoder architecture with ResNet 34 as the base encoder. BASNet (Bai et al, 2021) also uses ResNet encoder with skip connection in the decoder.…”

“…Sen1Floods11 Otsu thresholding (Bonafilia et al, 2020) 5 4 :58 Sen1Floods11 Sentinel-2 weak-label model (Bonafilia et al, 2020) 6 6 :21 BASNet (Bai et al, 2021) 5 3 :90 AN-34 (Helleis et al, 2022) 4 9 :70 S-1FS (Helleis et al, 2022) 5 4 :90 Cross-modal distillation (ours) 72:74…”

“…Sen1Floods11 (Bonafilia et al, 2020) is a popular dataset for flood segmentation, which was used in prior work that takes a single Sentinel-1 image as input (Katiyar et al, 2021; Ghosh et al, 2022; Helleis et al, 2022). It is publicly available and has a small set of high-quality hand-labeled images and a larger set of weak-labeled images.…”

Flood Detection with SAR: A Review of Techniques and Datasets

“…The Sen1Floods11 dataset has been exploited in [196] to test a modified UNet architecture and to investigate the fusion of S1 and S2 data. However, it is worth noting that Bereczky et al [197] reported issues of this dataset related to spatial autocorrelation of the samples.…”

Flood Modeling and Prediction Using Earth Observation Data

“…This success can be largely attributed to the generalizing power of CNNs toward spatiotemporal patterns in the data, specifically including contextual signatures, closely mimics human interpretation given enough training data. CNNs have already been successfully applied to various problems in the earth observation domain, including water and flood mapping using either optical or radar data (e.g., Bentivoglio et al 2021;Wieland and Martinis 2019;Li et al 2019a, b;Bonafilia et al 2020;Nemni et al 2020;Katiyar et al 2021;Bai et al 2021;Helleis et al 2022).…”