基于空间聚类种子生长算法的阵列干涉SAR点云滤波

王松; 张福博; 陈龙永; 梁兴东

doi:10.12000/JR18006

基于空间聚类种子生长算法的阵列干涉SAR点云滤波

DOI: 10.12000/JR18006

王松^1,2,,
张福博^1,,
陈龙永^1,2,,
梁兴东^1,2, ,

①.
中国科学院电子学研究所微波成像技术重点实验室北京 100190
②.
中国科学院大学北京 100049

基金项目: 国家部委基金

详细信息

作者简介:
陈龙永(1979–)，男，研究员，硕士生导师，现任中国科学院电子学研究所微波成像技术重点实验室常务副主任，主要从事高分辨率合成孔径雷达系统、干涉合成孔径雷达系统、微波成像新概念、新体制和新技术等领域的研究工作。E-mail: lychen@mail.ie.ac.cn

通讯作者:
梁兴东 xdliang@mail.ie.ac.cn

计量
- 文章访问数: 3163
- HTML全文浏览量: 560
- PDF下载量: 270
- 被引次数: 0
出版历程
- 收稿日期: 2018-01-19
- 修回日期: 2018-03-20
- 网络出版日期: 2018-06-28

Array-interferometric Synthetic Aperture Radar Point Cloud Filtering Based on Spatial Clustering Seed Growth Algorithm

Wang Song^{1,2
,},
Zhang Fubo^1
,,
Chen Longyong^{1,2
,},
Liang Xingdong^{1,2
, ,}

①.
Science and Technology on Microwave Imaging Laboratory, Institute of Electronics, Chinese Academy of Sciences, Beijing 100190, China
②.
University of Chinese Academy of Sciences, Beijing 100049, China

Funds: The National Ministries Foundation

摘要

摘要: 阵列干涉合成孔径雷达(Synthetic Aperture Radar, SAR)通过在交轨向布置多个天线，结合方位向的合成孔径和斜距向的大带宽信号，具备了3维分辨能力，且多个阵元保证了其在高程向的空间采样，能够解决干涉SAR(Interferometric SAR, InSAR)测绘中的叠掩问题，实现观测场景的3维成像。但是获得场景区域的3维点云分布中存在较多杂点，高程向误差较大，所以传统的激光雷达(Light Detection And Ranging, LiDAR)点云滤波方法不适用于阵列干涉SAR点云的滤波处理。针对该问题，该文提出基于空间聚类种子生长算法的阵列干涉SAR点云滤波算法，应用密度和高程双重阈值生成密度-高程图像，通过图像处理手段去除小型杂点，利用空间聚类种子生长算法将植被等从点云数据中去除，完成点云滤波处理。利用国内首次机载阵列干涉SAR实验数据，通过与传统LiDAR滤波方法进行比较，验证了该文算法的有效性，为后续建筑物提取和精细化处理提供保障。
- 阵列干涉SAR /
- 点云滤波 /
- 高程图像 /
- 空间聚类种子生长
Abstract: By arranging multiple antennas in the intersection direction and combining the synthetic aperture of azimuth direction and large bandwidth signal with oblique distance, array-interferometric Synthetic Aperture Radar (SAR) can generate a three-dimensional resolution and ensure the elevation spacial sampling due to its multiple array element, which could avoid the layover problem in surveying and mapping in the Interference SAR (InSAR) and realize the three-dimensional imaging of the observation scene. However, considering the existence of too much noise in the three-dimensional point cloud distribution in the scene area and the large elevation error, the traditional Light Detection And Ranging (LiDAR) point cloud filtering method is not suitable for the filtering processing of the array-interferometric SAR point cloud. In order to solve this problem, an array-interferometric SAR point cloud filtering algorithm based on spatial clustering seed growth algorithm is proposed, in which the density-elevation image is generated by the double threshold of density and elevation, the small clutter is removed by image processing, and the vegetation is removed from the point cloud data by using the spatial clustering seed growth algorithm, thus the point cloud filtering process is completed. Using the first airborne array-interferometric SAR experimental data, the validity of the proposed algorithm is verified compared to the traditional LiDAR filtering method, which provides the guarantee for the subsequent building extraction and meticulous treatment.
- Array-interferometric synthetic aperture radar /
- Point cloud filtering /
- Elevation image /
- Spatial clustering seed growth

HTML全文

图 1 原始3维点云

Figure 1. The original three-dimensional point cloud

下载: 全尺寸图片幻灯片

图 2 简单建筑物模型

Figure 2. Simple building model

下载: 全尺寸图片幻灯片

图 3 搜索中邻域的编码

Figure 3. The encoding of the neighborhood in the search

下载: 全尺寸图片幻灯片

图 4 种子生长流程图

Figure 4. Flowchart of seed growth

下载: 全尺寸图片幻灯片

图 5 点云处理前后示意图

Figure 5. Schematic diagram before and after cloud processing

下载: 全尺寸图片幻灯片

图 6 处理后局部放大图

Figure 6. Local enlarged image after processing

下载: 全尺寸图片幻灯片

图 7 图像处理前后的密度-高程图像

Figure 7. Density-elevation image before and after image processing

下载: 全尺寸图片幻灯片

图 8 图像处理前后的局部放大图

Figure 8. Local enlargement before and after image processing

下载: 全尺寸图片幻灯片

图 9 小面积区域去除前后的密度-高程图像

Figure 9. Density-elevation images before and after removal of small area

下载: 全尺寸图片幻灯片

图 10 滤波后结果与光学图像对照图

Figure 10. Comparison of filter results with optical image

下载: 全尺寸图片幻灯片

表 1 滤波算法质量评价表

Table 1. Filter algorithm quality evaluation table

滤波算法	After filtering	T_p	F_p	F_N	Completeness (100%)	Correctness (100%)	Quality (100%)
本文算法	64858	63753	1105	2381	96.40	98.30	94.81
形态学滤波	81443	64032	17441	2102	96.82	78.62	76.64
坡度滤波	84039	64395	19944	1739	97.37	76.63	75.07

下载: 导出CSV

参考文献(16)

[1]	Cumming I G and Wong F H著. 洪文, 胡东辉, 译. 合成孔径雷达成像-算法与实现[M]. 北京: 电子工业出版社, 2007: 3–5. Cumming I G and Wong F H. Hong Wen and Hu Dong-hui, Trans. Digital Processing of Synthetic Aperture Radar Data: Algorithms and Implementation[M]. Beijing: Publishing House of Electronics Industry, 2007: 3–5.
[2]	Dorninger P and Pfeifer N. A comprehensive automated 3D approach for building extraction, reconstruction, and regularization from airborne laser scanning point clouds[J]. Sensors, 2008, 8(11): 7323–7343. DOI: 10.3390/s8117323
[3]	Filin S. Surface clustering from airborne laser scanning data[J]. International Archives of Photogrammetry and Remote Sensing, 2002, 32(3A): 119–124.
[4]	Elaksher A F and Bethel J S. Reconstructing 3D buildings from Lidar data[C]. Proceedings of the ISPRS Commission III Symposium on Photogrammetric Computer Vision, Graz, Austria, 2002: 1–6.
[5]	Forlani G, Nardinocchi C, Scaioni M, et al. Complete classification of raw LIDAR data and 3D reconstruction of buildings[J]. Pattern Analysis and Applications, 2006, 8(4): 357–374. DOI: 10.1007/s10044-005-0018-2
[6]	Zhu X X and Shahzad M. Facade reconstruction using Multiview Spaceborne TomoSAR point clouds[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(6): 3541–3552. DOI: 10.1109/TGRS.2013.2273619
[7]	Shahzad M and Zhu X X. Reconstructing 2-D/3-D building shapes from spaceborne Tomographic Synthetic Aperture Radar data[C]. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XL-3, 2014, ISPRS Technical Commission III Symposium, Zurich, Switzerland, 2014.
[8]	Shahzad M and Zhu X X. Automatic detection and reconstruction of 2-D/3-D building shapes from Spaceborne TomoSAR point clouds[J]. IEEE Transactions on Geoscience and Remote Sensing, 2016, 54(3): 1292–1310. DOI: 10.1109/TGRS.2015.2477429
[9]	Sithole G and Vosselman G. Experimental comparison of filter algorithms for bare-earth extraction from airborne laser scanning point clouds[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2004, 59(1/2): 85–101. DOI: 10.1016/j.isprsjprs.2004.05.004
[10]	Vosselman G. Slope based filtering of laser altimetry data[J]. International Archives of Photogrammetry and Remote Sensing, 2000, 33(B3/2): 935–942.
[11]	黄先锋, 李卉, 王潇, 等. 机载LiDAR数据滤波方法评述[J]. 测绘学报, 2009, 38(5): 466–469 Huang Xian-feng, Li Hui, Wang Xiao, et al. Filter algorithms of airborne LiDAR data: Review and prospects[J]. Acta Geodaetica et Cartographica Sinica, 2009, 38(5): 466–469
[12]	Zhu X X and Bamler R. Super-resolution power and robustness of compressive sensing for spectral estimation with application to spaceborne tomographic SAR[J]. IEEE Transactions on Geoscience and Remote Sensing, 2012, 50(1): 247–258. DOI: 10.1109/TGRS.2011.2160183
[13]	Lombardini F. Differential tomography: A new framework for SAR interferometry[J]. IEEE Transactions on Geoscience and Remote Sensing, 2005, 43(1): 37–44. DOI: 10.1109/TGRS.2004.838371
[14]	Shahzad M and Zhu X X. Robust reconstruction of building facades for large areas using spaceborne TomoSAR point clouds[J]. IEEE Transactions on Geoscience and Remote Sensing, 2015, 53(2): 752–769. DOI: 10.1109/TGRS.2014.2327391
[15]	栾晓岩. 一种TIN生成算法及其三维显示[J]. 海洋测绘, 2004, 24(5): 39–41. DOI: 10.3969/j.issn.1671-3044.2004.05.012 Luan Xiao-yan. The method of building TIN and its three-dimensional display[J]. Hydrographic Surveying and Charting, 2004, 24(5): 39–41. DOI: 10.3969/j.issn.1671-3044.2004.05.012
[16]	李杭, 梁兴东, 张福博, 等. 基于高斯混合聚类的阵列干涉SAR三维成像[J]. 雷达学报, 2017, 6(6): 630–639. DOI: 10.12000/JR17020 Li Hang, Liang Xing-dong, Zhang Fu-bo, et al. 3D imaging for array InSAR based on Gaussian mixture model clustering[J]. Journal of Radars, 2017, 6(6): 630–639. DOI: 10.12000/JR17020