圆周扫描地基SAR地形表面提取辅助的三维成像算法

Three-dimensional Imaging Algorithm Assisted by Terrain Surface Extraction for Circular Scanning Ground-based SAR

圆周扫描地基合成孔径雷达(GBCSAR)具备大视场、高分辨率等优势,且具备三维分辨能力,在复杂场景形变监测领域具有重要意义。但在大场景下目标点数量庞大,常规的三维成像方法耗时很长,难以满足实时连续监测的需求。因此本文提出一种针对GBCSAR的地形表面成像方法。在场景形变监测时,地形表面形变往往可以反映场景形变趋势,针对这一特性,本文方法将借助地形表面实现成像。利用地形表面提取得到的有效点曲面代替场景的三维空间目标点,实现了降维成像,提高了成像效率。本文首先给出了GBCSAR的几何模型以及信号模型。然后论述了地表三维曲面的提取方法和原理,之后论述了三维曲面成像算法原理。最后利用仿真实验验证算法的可行性,与原始数据图像进行比较分析,并与传统三维成像算法的成像效率进行比较。

GBCSAR has the advantages of large field of view,high resolution,and three-dimensional resolution capabilities.It is of great significance in the field of complex scene deformation monitoring.However,there are a large number of target points in a large scene,and the traditional three-dimensional imaging method takes a long time,which is difficult to meet the demand for continuous monitoring.Therefore,this paper proposes a terrain surface imaging algorithm for GBCSAR.In scene deformation monitoring,terrain surface deformation can often reflect the trend of scene deformation.Aiming at this characteristic,the method in this paper will use the terrain surface to achieve imaging.The effective point surface obtained by the terrain surface extraction is used to replace the three-dimensional space target point of the scene,which realizes the dimensionality reduction imaging and improves the imaging efficiency.First,the geometric model and signal model of GBCSAR are given.Then it discusses the extraction method of the three-dimensional surface of the terrain,and then discusses the principle of the three-dimensional surface imaging algorithm.Finally,the feasibility of the method is verified through simulation experiments,and the imaging efficiency is compared with traditional three-dimensional imaging methods.