基于DEM的SAR图像洪水水体的提取

Extracting the flood extent from SAR imagery on basis of DEM

星载雷达遥感已广泛应用于洪水灾害的监测和评估中。但是由于其阴影与水体严重混淆 ,使得从雷达图像中提取洪水水体较为困难。为此 ,本文探讨了在地形数据的支持下 ,实现星载SAR图像洪水水体半自动提取的方法。主要包括以下几步 :首先 ,对雷达图像进行滤波处理 ,并将其与地形数据匹配 ,从雷达图像中提取出初步的洪水水体范围 ;其次 ,利用地形数据生成DEM ,并根据雷达图像的成像参数和DEM生成相应的模拟雷达图像 ;最后 ,利用模拟雷达图像上的阴影 ,剔除被误提为洪水水体的阴影 ,从而实现洪水水体的准确提取。研究表明 ,该方法能有效地、半自动地将星载雷达图像上的洪水水体提取出来 ;同时 ,该方法可以用于多种星载多模式的雷达图像的洪水水体的识别提取 ,尤其对含山区的大范围洪水水体的半自动提取 ,更为适用。

Satellite SAR imagery is broadly used in monitoring and evaluating flood disaster. It is difficult to automatically extract the flood extent from RADARSAT SAR imagery because of the confusion in imagery gray of flood and shadow. Therefore, extracting the flood extent from Satellite SAR imagery with the support of topographic data is explored here. It includes several steps as follows. Firstly, RADARSAT SAR imagery is filtered by enhanced frost filter with window size of 7 pixels, and geo registered to topographic data. Secondly, flood extent is primarily extracted from RADARSAT SAR imagery. Thirdly, DEM is created from topographic data by using GIS software. Fourthly, simulated SAR imagery is created from DEM. Finally, the simulated SAR imagery is registered to RADARSAT SAR imagery, the shadow from the simulated SAR imagery is used to mask the mislabeled flood extent from RADARSAT SAR due to its shadow influence. It is shown that the flood extent can be accurately, semi automatically extracted from RADARSAT SAR with the method,and that the method is suitable not only for RADARSAT SAR imagery but also for other satellite SAR imagery with different imaging modes, specially for SAR imagery covering mountainous area.