摘要
星载合成孔径雷达(SAR)直接定位算法是发展SAR影像椭球表面校正地理编码(GEC)和地形校正地理编码(GTC)算法的基础。发展高精度的直接定位算法对促进SAR影像的应用具有重要意义。本文在距离-多普勒(RD)定位模型定位算法的基础上,发展了一种将纯分析法(AGM)和纯数值计算法相结合的算法(AIRGM)。新的定位算法兼有纯分析法和纯数值计算法的优点。利用一景ERS-2 SAR SLC数据,采用两种精度评价方法,将所发展的算法与现有的两种算法的定位精度进行了对比。第1种精度评价方法以影像产品中提供的四角点大地坐标为参考,第2种方法以从1∶5万地形图上获取的14个控制点为参考。评价结果表明AIRGM算法不仅定位精度较高,而且执行效率比纯粹的数值解算法(ASF)有了很大的提高。
The geo-location algorithm of Space-borne Synthetic Aperture Radar（SAR） is the fundamental for developing geocoded-ellipsoid correction（GEC） and geocoded-terrain correction （GTC） algorithm, so it is very important to develop high accuracy geo-location algorithm for the application of SAR imagery. After detailed investigation of the currently available SAR geo-location algorithms, a new geo-location methodology AIRGM （Analysis and Iterating Routes Based Geolocation Method） was developed. AIRGM was developed from the pure analysis geo-location method （AGM） by integrating numerical iteration routes into AGM. So this new geo-location methodology has advantages of both pure analysis method（efficient） and pure numerical method（high geolocation accuracy）. Applying the three geolocation methods to one scene of ERS-2 SAR SLC imagery, the geolocation accuracy of AIRGM was compared with the other two published geo- location methodologies （ASF and AGM） through two kinds of validation method. The first validation method takes the geographic coordinate of the four image corner points provided with SAR data as reference, while the second method takes 14 ground control points（GCPs） collected from 1：50 000 geographical map sheets as reference. It has been shown that the new geo-location algorithm can achieve higher geo-location accuracy, and the efficiency of geo-location processing is much higher than the pure numerical geo-location method（ASF） , which is built upon many iteration routes.
出处
《中国图象图形学报》
CSCD
北大核心
2006年第8期1105-1109,共5页
Journal of Image and Graphics
基金
国家"863"项目(2005AA132030)
