参考地理数据的大比例尺影像无控定位

Uncontrolled positioning of large-scale image with reference to geographic data

针对传统摄影测量影像定位通过人工方式获取控制点导致的效率低、费用高等问题,本文提出了一种参考地理数据的大比例尺影像无控定位方法。首先利用经典SIFT算法从参考地理数据(DEM和DOM)提取带有控制信息的特征点;然后将通过POS获取的DSM配准到DEM地理坐标上,并利用形态学滤波对DSM进行处理,以提取并剔除建筑物、树木等高程剧烈变化区域的匹配点,降低校正误差;最后进行影像校正和定位。试验结果表明,本文方法能够满足大比例尺航空影像生产定位精度要求,具有高效生产、节约资源等优势。

Aiming at the problems of low efficiency and high cost of acquiring control points manually by traditional photogrammetric image positioning,this paper proposes a method for uncontrolled positioning of large-scale images with reference geographic data.Firstly,it uses the classic SIFT algorithm to extract feature points with control information from the reference geographic data(DEM and DOM).Secondly,it registers DSM obtained through POS to geographic coordinates of DEM,which is morphologically filtered in order to extract and remove the matching points in areas where the elevation changes drastically due to buildings,trees and so on,so as to reduce the correction error.Finally,it corrects and locates the image.The experiments show that the proposed method in this paper can meet the positioning accuracy requirements of large-scale aerial image production.Also,this method shows the advantages of efficient production and resource saving.