Near real-time flood detection in rural and urban areas using high resolution Synthetic Aperture Radar images

洪水是世界范围内农村和城市地区的主要灾害，最近在英国经常发生。一种近乎实时的洪水检测算法可以对城市和农村地区的洪水程度进行概览，即使有云存在，也可以在夜间和白天工作，这可能是一种有用的洪水救援管理工具。最新一代的高分辨率合成孔径雷达（SAR）卫星现在使这种技术成为可能。绝大多数被洪水淹没的地区可能是农村而不是城市，但探测城市洪水是很重要的，因为与之相关的风险和成本都在增加。可以使用ERS和ASAR等sar来检测农村洪水的洪水程度，但这些分辨率太低（25米），无法检测城市地区被洪水淹没的街道。然而，最近已经发射了一些空间分辨率高达1米的sar，能够探测城市洪水。它们包括TerraSAR-X、RADARSAT-2、ALOS PALSAR和cosmos - skymed的前三颗卫星。使接近实时操作成为可能的一个重要因素是，现在可以快速进行精确的地理定位。例如，利用对轨道参数的精确了解，TerraSAR-X的图像可以以地理注册的形式提供，定位精度超过一个像素。在没有明显的风或雨的情况下，河流洪水通常在SAR图像中显示为黑色，因为水起到了镜面反射的作用。DLR Oberpfaffenhofen的卫星危机信息中心已经实施了一种近乎实时的洪水检测算法，该算法使用基于分裂的自动阈值处理程序，应用于多视单极化TerraSAR-X数据。该方法使用区域增长迭代分割/分类方法在低SAR反向散射区域搜索水，并且需要最小的用户干预。然而，该算法需要修改才能在包含雷达阴影和中途停留的城市地区工作。相比之下，以前还开发了一种使用TerraSAR-X检测城市地区洪水的半自动算法。它使用DLR SAR端到端模拟器（SETES）结合LiDAR数据来估计由于雷达阴影或建筑物和较高植被造成的中途停留而看不到水的图像区域。该算法旨在检测洪水范围，以便在离线情况下校准和验证城市洪水淹没模型，并且需要用户在多个阶段进行交互。这必然会给最终产品的生产带来延迟的因素。该建议是修改和结合现有算法，使需要人工交互的步骤自动化，并利用城市地区激光雷达数据的可用性，从而实现接近实时的算法。这将在2007年图克斯伯里洪水的TerraSAR-X图像上进行测试。

Flooding is a major hazard in both rural and urban areas worldwide, and has occurred regularly in the UK in recent times. A near real-time flood detection algorithm giving a synoptic overview of the extent of flooding in both urban and rural areas, and capable of working during night-time and day-time even if cloud was present, could be a useful tool for operational flood relief management. The latest generation of very high resolution Synthetic Aperture Radar (SAR) satellites now make such technology a real possibility. The vast majority of a flooded area may be rural rather than urban, but it is important to detect the urban flooding because of the increased risks and costs associated with it. Flood extent can be detected in rural floods using SARs such as ERS and ASAR, but these have too low a resolution (25m) to detect flooded streets in urban areas. However, a number of SARs with spatial resolutions as high as 1m have recently been launched that are capable of detecting urban flooding. They include TerraSAR-X, RADARSAT-2, ALOS PALSAR, and the first three of the COSMO-SkyMed satellites. An important factor making near real-time operation possible is that accurate geo-registration can now be performed rapidly. For example, the images from TerraSAR-X can be made available in geo-registered form to better than one pixel locational accuracy using precise knowledge of the orbit parameters. In the absence of significant wind or rain, river flood-water generally appears dark in a SAR image because the water acts as a specular reflector. A near real-time flood detection algorithm using a split-based automatic thresholding procedure applied to multi-look single-polarisation TerraSAR-X data has been implemented at DLR Oberpfaffenhofen's Centre for Satellite-Based Crisis Information. This searches for water as regions of low SAR backscatter using a region-growing iterated segmentation/classification approach, and requires minimal user intervention. However, the algorithm would require modification to work in urban areas containing radar shadow and layover. In contrast, a semi-automatic algorithm for the detection of floodwater in urban areas using TerraSAR-X has also been developed previously. It uses the DLR SAR End-To-End simulator (SETES) in conjunction with LiDAR data to estimate regions of the image in which water would not be visible due to radar shadow or layover caused by buildings and taller vegetation. The algorithm is aimed at detecting flood extents for calibrating and validating an urban flood inundation model in an offline situation, and requires user interaction at a number of stages. This invariably introduces an element of delay into the production of the final product. The proposal is to revise and combine the existing algorithms to automate the steps requiring manual interaction and to take advantage of the availability of LiDAR data in the urban area, to lead to a near real-time algorithm. This would be tested on the TerraSAR-X image of the Tewkesbury 2007 flood.