地形特征约束下的失真DEM修复方法

发布时间：2018-03-10 02:21

本文选题：数字高程模型　切入点：地形特征　出处：《测绘通报》2017年08期 　论文类型：期刊论文

【摘要】：数字高程模型(DEM)在描述地形时会存在一定程度的失真现象,对失真DEM进行修复可以有效地提高其精度及保真度。本文针对DEM在描述规则地形时存在的失真现象,提出了用地形特征约束失真DEM形态结构的失真修复方法,以规则地形中的平地结构地形为例,进行了失真修复试验。首先根据地形特征确定平地地形的DEM应为平面结构,再采用最小二乘法(LS)及随机抽样一致性算法(RANSAC)对其平面方程进行参数估计,进而得到修复的DEM数据。试验结果表明,本文所提的修复方法有效可行,修复后的DEM数据不仅提高了精度,还能充分地反映原地形的地形特征。两种修复算法的修复能力相接近,当存在大量格网点时,都能达到很好的修复效果,但RANSAC更能适应高程异常的情况,鲁棒性更好。
[Abstract]:Digital elevation model (DEM) will have a certain degree of distortion in describing terrain, and repairing distorted DEM can effectively improve its accuracy and fidelity. This paper aims at the distortion phenomenon of DEM in describing regular terrain. In this paper, a distortion restoration method of distorted DEM morphological structure with terrain feature constraint is proposed. Taking flat terrain in regular terrain as an example, distortion restoration experiments are carried out. Firstly, it is determined that the DEM of flat terrain should be planar structure according to terrain characteristics. Then the least square method and random sampling consistency algorithm are used to estimate the parameters of the plane equation, and the repaired DEM data are obtained. The experimental results show that the proposed method is effective and feasible. The restored DEM data not only improves the accuracy, but also fully reflects the topographic features of the original terrain. But RANSAC can adapt to the abnormal elevation better robustness.
【作者单位】：安徽省智慧城市与地理国情监测重点实验室;合肥工业大学;
【基金】：安徽省智慧城市与地理国情监测重点实验室开放性课题基金(2016-K-05Z)
【分类号】：P208;P217

【相似文献】