摘要
利用正交双航过的全极化合成孔径雷达(SAR)图像,采用圆极化方法计算了观测区域在两次独立观测时的极化倾角,计算得到观测区域内的二维坡度分布图。由于SAR图像特有的speckle噪声、阴影和几何形变等,二维坡度图中存在大量与实际坡度值不符的"奇异点",对此提出一种"奇异点"搜索算法,标记出二维坡度图中的"奇异点",并采用分形插值的方法重建了"奇异点"处的坡度。最后利用加权完整多重网格法(WFMG)重建了地形高度,和原始DEM数据的对比验证了该方法的有效性。
Using full polarimetric SAR images of two cross track, polarization orientation angles are obtained by circular polarization method. 2-D terrain slope map is generated from polarization orientation angles. Because of speckle noise, shadows, layover of SAR images, there are many irregular points in terrain slope map. A method is implemented to search these irregular points. After that, fractal interpolation is used to estimate the slopes of the irregular points. Terrain height is reconstructed by Weighted full multigrid (WFMG) algorithm.
出处
《信号处理》
CSCD
北大核心
2009年第1期48-51,共4页
Journal of Signal Processing
关键词
极化
合成孔径雷达
分形插值
测高
Polarization
Synthetic Aperture Radar
Fractal interpolation
Terrain mapping