基于有理多项式模型的星载InSAR影像制作数字高程模型的研究

Research of DEM Generation by Spaceborne InSAR Images Based RFM Model

合成孔径雷达干涉测量是一种利用SAR影像复数据的相位信息提取地面三维信息的技术,它通过获取地面目标在两幅SAR影像上的相位差值来解算该点的三维坐标。有理多项式模型作为一种数学意义上的几何模型,独立于传感器和平台,简单且具有通用性,可建立地面任意坐标与影像空间的关系。本文在RFM模型用于替代星载SAR的距离多普勒模型和星载InSAR的干涉相位方程基础上,研究RFM模型应用于星载高分辨率InSAR影像制作DEM的可行性和精度。利用兰州COSMOSkyMed数据和资源三号三线阵数据制作的DEM为参考数据进行试验验证,与资源三号制作的DEM结果相比,RFM模型用于InSAR技术生成的DEM中误差为7.55m。

Synthetic aperture radar interferometry （InSAR） is a new technique by phase data of SAR complex images for extracting the terrain elevation information, which can solve geographic coordinates of ground points by phase difference of two SAR images. RFM model isa mathematical sense of the imaging geometry model, which is independent of the sensor and platform. This model can establish the relationship between the ground coordinates of the image and the corre- sponding geographic coordinates. At present, some work has been studied about RFM model as a replacement for range-doppler model and phase equation, based on that, this paper states how RFM model is applied for DEM generation by InSAR technique, and then RFM model can be used in four steps, coarse registration, subtract flatten phase, phase height conversion and geocoding. Based on the contrast experiment between COSMO-SkyMed SAR images and ZY-3 optical images for Lanzhou area, mean square error of DEM generated by the RFM model applied for InSAR technology is 7.55 meters.