UrbanSARFloods: Sentinel-1 SLC-Based Benchmark Dataset for Urban and Open-Area Flood Mapping

Jie Zhao,Zhitong Xiong,Xiao Xiang Zhu

2024-06-06

Abstract:Due to its cloud-penetrating capability and independence from solar illumination, satellite Synthetic Aperture Radar (SAR) is the preferred data source for large-scale flood mapping, providing global coverage and including various land cover classes. However, most studies on large-scale SAR-derived flood mapping using deep learning algorithms have primarily focused on flooded open areas, utilizing available open-access datasets (e.g., Sen1Floods11) and with limited attention to urban floods. To address this gap, we introduce \textbf{UrbanSARFloods}, a floodwater dataset featuring pre-processed Sentinel-1 intensity data and interferometric coherence imagery acquired before and during flood events. It contains 8,879 $512\times 512$ chips covering 807,500 $km^2$ across 20 land cover classes and 5 continents, spanning 18 flood events. We used UrbanSARFloods to benchmark existing state-of-the-art convolutional neural networks (CNNs) for segmenting open and urban flood areas. Our findings indicate that prevalent approaches, including the Weighted Cross-Entropy (WCE) loss and the application of transfer learning with pretrained models, fall short in overcoming the obstacles posed by imbalanced data and the constraints of a small training dataset. Urban flood detection remains challenging. Future research should explore strategies for addressing imbalanced data challenges and investigate transfer learning's potential for SAR-based large-scale flood mapping. Besides, expanding this dataset to include additional flood events holds promise for enhancing its utility and contributing to advancements in flood mapping techniques.

Computer Vision and Pattern Recognition,Image and Video Processing

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is that the existing large - scale Synthetic Aperture Radar (SAR) flood mapping research mainly focuses on floods in open areas, while having limited attention to urban floods. Specifically, most of the existing studies have used publicly available data sets (such as Sen1Floods11), which mainly focus on floods in open areas and are less involved in mapping urban floods. This has led to obvious deficiencies in urban flood detection, especially in dealing with the problems of data imbalance and small - scale training data sets.

To fill this gap, the author introduced a new data set - UrbanSARFloods, which contains pre - processed Sentinel - 1 intensity data and interferometric coherence images, covering data before and after flood events. This data set contains 8,879 512×512 chips, covering 807,500 square kilometers, across 20 surface cover types and 5 continents, with a total of 18 flood events. Through this data set, the author evaluated the performance of the existing state - of - the - art Convolutional Neural Networks (CNNs) in segmenting open - area and urban flood areas.

The study found that existing methods, including the use of weighted cross - entropy (WCE) loss and transfer learning, although improving performance to a certain extent, still have difficulty overcoming the challenges brought by data imbalance and small - scale training data sets. Urban flood detection remains a difficult problem, and future research needs to explore strategies to solve the data imbalance problem and further explore the potential of transfer learning in large - scale SAR - based flood mapping.

In addition, the author also pointed out that expanding the data set to include more flood events is expected to enhance its utility and promote the development of flood mapping technology. The UrbanSARFloods data set, including training, validation data and raw data, can be found on GitHub (https://github.com/jie666 - 6/UrbanSARFloods).