自适应局部空谱一致下的机载LiDAR数据建筑物提取算法

doi:10.11947/j.AGCS.2024.20230517

摘要/Abstract

摘要：

现有研究均将建筑物激光反射强度的全局统计特性用于辅助机载激光雷达数据的建筑物提取，但是这类方法无法满足大尺度复杂城区场景下光谱迥异建筑物完备、准确提取的需求。为此，本文提出一种自适应局部空谱一致下的机载LiDAR数据建筑物提取方法。该方法首先将原始机载LiDAR数据转换为LiDAR 3D影像。然后，依据建筑物高程跳变和边缘局部接近直线特性选取种子体素，进而依据单体建筑示出的空间连续性及光谱一致性将与种子体素空谱一致的连通分量标记为建筑物屋顶。其中，单体建筑的光谱一致性由统计空间连通的种子体素的强度特性给出。最后，结合建筑屋顶对立面的空间约束以及立面的局部强度一致性约束提取建筑物立面。该方法通过自适应于各建筑单体光谱的设计解决了不符合光谱全局统计特性的建筑物的准确提取问题，提升了点云光谱信息的使用价值，并由此拓宽了点云光谱信息的应用场景。选用国际摄影测量与遥感协会提供，不同复杂程度的3组城区场景实测机载LiDAR数据测试本文方法的可行性和有效性。试验结果表明：可实现不同复杂程度场景下的建筑物提取；建筑物提取结果的平均完整率、正确率及质量分别为99.0%、98.0%、96.8%，明显优于传统的利用光谱全局统计特性的建筑物提取方法。

关键词: 机载激光雷达, 建筑物提取, 空谱一致, 点云, 全局统计特性

Abstract:

All the existing studies use the global statistical characteristics of laser reflection intensity of buildings to aid the extraction of buildings from airborne LiDAR data, but this solution cannot meet the needs of comprehensive and accurate extraction of buildings with different spectra in large-scale complex urban scenes. Therefore, a building extraction method from airborne LiDAR data based on adaptive local spatial-spectral consistency is developed. The proposed method first converts raw airborne LiDAR data into 3D image. Then, the seeds are selected according to the characteristics of building elevation jump and edge approaching straight line. Subsequently, the connected components that are spatially and spectrally consistent with the seedsare labeled as the building roof, in which the spectral consistency is given by the statistical intensity properties of an individual building. Finally, the building facade is extracted by combining the spatial constraint of the extracted building roof and the local intensity consistency constraints. This method solves the problem of accurate extraction of buildings that do not conform to the global statistical characteristic of the spectrum by self-adapting to the local spectrum of each individual building, improves the use value of point cloud spectral information, and thus broadens the application scenarios of point cloud spectral information. Three airborne LiDAR datasets of urban scene with different complexities provided by International Association for Photogrammetry and Remote Sensing are used to test the feasibility and effectiveness of the proposed method. The experimental results show that the proposed method can excellently extract buildings in scenes with different complexities. The average completeness, accuracy and quality of the building extraction results are 99.0%, 98.0% and 96.8%, respectively, which are obviously better than the traditional building extraction method using the global statistical properties of the spectrum.

Key words: airborne LiDAR, building extraction, spatial-spectral consistency, point cloud, global statistical characteristic

中图分类号:

P236

王丽英, 张康丽, 李鑫奥, 有泽, 丰勇. 自适应局部空谱一致下的机载LiDAR数据建筑物提取算法[J]. 测绘学报, 2024, 53(12): 2349-2360.

Liying WANG, Kangli ZHANG, Xinao LI, Ze YOU, Yong FENG. An algorithm for building extraction from airborne LiDAR data under adaptive local spatial-spectral consistency[J]. Acta Geodaetica et Cartographica Sinica, 2024, 53(12): 2349-2360.

图/表 15

图1

图2

图3

图4

图5

图6

图7

表1

图8

图9

图10

表2

图11

图12

表3

参考文献 31

[1]	杜建丽, 陈动, 张振鑫, 等. 建筑点云几何模型重建方法研究进展[J]. 遥感学报, 2019, 23(3): 374-391.
	DU Jianli, CHEN Dong, ZHANG Zhenxin, et al. Research progress of building reconstruction via airborne point clouds[J]. Journal of Remote Sensing, 2019, 23(3): 374-391.
[2]	ELBERINK S O, MAAS H G. The use of anisotropic height texture measures for the segmentation of laser scanner data[J]. The International Archives of The Photogrammetry Remote Sensing and Spatial Information Sciences, 2000, 33(B3/2): 678-684.
[3]	尤红建, 苏林, 李树楷. 利用机载三维成像仪的DSM数据自动提取建筑物[J]. 武汉大学学报(信息科学版), 2002, 27(4): 408-413.
	YOU Hongjian, SU Lin, LI Shukai. Automatic extraction of buildings from DSM acquired by airborne three-dimensional imager[J]. Geomatics and Information Science of Wuhan University, 2002, 27(4): 408-413.
[4]	李泽刚. 基于机载LiDAR数据与影像处理技术的建筑物提取研究[D]. 南昌: 东华理工大学, 2015.
	LI Zegang. Study on the extraction of buildings based on the airborne LiDAR data and image processing technology[D]. Nanchang: East China University of Technology, 2015.
[5]	陈永枫. 基于机载LiDAR点云数据的建筑物重建技术研究[D]. 郑州: 信息工程大学, 2013.
	CHEN Yongfeng. Research on airborne LiDAR points cloud data building reconstruction technology[D]. Zhengzhou: Information Engineering University, 2013.
[6]	郭佳. 机载LiDAR点云数据滤波及建筑物提取技术研究[D]. 西安: 长安大学, 2014.
	GUO Jia. Study on data filtering and buildings abstraction of airborne LiDAR points cloud data[D]. Xi'an: Changan University, 2014.
[7]	WANG Ruisheng, HU Yong, WU Huayi, et al. Automatic extraction of building boundaries using aerial LiDAR data[J]. Journal of Applied Remote Sensing, 2016, 10(1): 016022.
[8]	VOSSELMAN G. Building reconstruction using planar faces in very high density height data[J]. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 1999, 32(3): 87-94.
[9]	ALBANO R. Investigation on roof segmentation for 3D building reconstruction from aerial LiDAR point clouds[J]. Applied Sciences, 2019, 9(21): 4674.
[10]	YI Zhenghui, WANG Haotong, DUAN Guangyao, et al. An airborne LiDAR building-extraction method based on the naive Bayes-RANSAC method for proportional segmentation of quantitative features[J]. Journal of the Indian Society of Remote Sensing, 2021, 49(2): 393-404.
[11]	LIU Maohua, SHAO Yue, LI Ruren, et al. Method for extraction of airborne LiDAR point cloud buildings based on segmentation[J]. PLoS One, 2020, 15(5): e0232778.
[12]	TAKANO K T, DOIHARA T, SHIBASAKI R. Automatic building extraction and 3-D city modeling from LiDAR data based Hough transformation[EB/OL]. [2023-09-10]. https://www.isprs.org/proceedings/XXXV/congress/comm3/papers/280.pdf.
[13]	CHEN Dong, ZHANG Liqiang, LI J, et al. Urban building roof segmentation from airborne LiDAR point clouds[J]. International Journal of Remote Sensing, 2012, 33(20): 6497-6515.
[14]	王丽英, 王圣, 徐艳, 等. 结合体元数据结构的机载LiDAR建筑物检测[J]. 中国图象图形学报, 2017, 22(10): 1436-1446.
	WANG Liying, WANG Sheng, XU Yan, et al. Airborne LiDAR building detection based on voxel data structure[J]. Journal of Image and Graphics, 2017, 22(10): 1436-1446.
[15]	WANG Liying, XU Yan, LI Yu. A voxel-based 3D building detection algorithm for airborne LiDAR point clouds[J]. Journal of the Indian Society of Remote Sensing, 2019, 47: 349-358.
[16]	BRUNN A, WEIDNER U. Hierarchical Bayesian nets for building extraction using dense digital surface models[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 1998, 53(5): 296-307.
[17]	ZHANG Zhenxin, ZHANG Liqiang, TONG Xiaohua, et al. Discriminative-dictionary-learning-based multilevel point-cluster features for ALS point-cloud classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2016, 54(12): 7309-7322.
[18]	YANG Juntao, KANG Zhizhong, AKWENSI P H. A label-constraint building roof detection method from airborne LiDAR point clouds[J]. IEEE Geoscience and Remote Sensing Letters, 2021, 18(8): 1466-1470.
[19]	WANG Liying, XU Yan, LI Yu, et al. Voxel segmentation-based 3D building detection algorithm for airborne LiDAR data[J]. PLoS One, 2018, 13(12): e0208996.
[20]	WICHMANN V, BREMER M, LINDENBERGER J, et al. Evaluating the potential of multispectral airborne LiDAR for topographic mapping and land cover classification[EB/OL]. [2023-09-10]. https://isprs-annals.copernicus.org/articles/II-3-W5/113/2015/isprsannals-II-3-W5-113-2015.pdf.
[21]	MORSY S, SHAKER A, EL-RABBANY A, et al. Airborne multispectral LiDAR data for land-cover classification and land/water mapping using different spectral indexes[EB/OL]. [2023-09-10]. https://isprs-annals.copernicus.org/articles/III-3/217/2016/isprs-annals-III-3-217-2016.pdf.
[22]	CHEN Biwu, SHI Shuo, GONG Wei, et al. Multispectral LiDAR point cloud classification: a two-step approach[J]. Remote Sensing, 2017, 9(4): 373.
[23]	王丽英, 有泽, 吴际, 等. 联合NDRI特征和空间相关性的机载MS-LiDAR数据分类[J]. 红外与激光工程, 2023, 52(2): 20220376.
	WANG Liying, YOU Ze, WU Ji, et al. Airborne MS-LiDAR data classification by combining NDRI features and spatial correlation[J]. Infrared and Laser Engineering, 2023, 52(2): 20220376.
[24]	王丽英, 吴际, 有泽, 等. 多维GMM与邻域约束的多光谱机载LiDAR数据城市地物分类[J]. 测绘学报, 2023, 52(3): 419-431. DOI:. doi: 10.11947/j.AGCS.2023.20210153
	WANG Liying, WU Ji, YOU Ze, et al. Urban object classification of multispectral airborne LiDAR data with multidimensional Gauss mixture model and neighborhood constraints[J]. Acta Geodaetica et Cartographica Sinica, 2023, 52(3): 419-431. DOI:. doi: 10.11947/j.AGCS.2023.20210153
[25]	HAGSTROM S T. Voxel-based LiDAR analysis and applications[D]. Rochester: Rochester Institute of Technology, 2014.
[26]	王丽英, 王鑫宁. 多值体素连通区域构建下的机载LiDAR数据三维平面提取[J]. 地球信息科学学报, 2021, 23(9): 1598-1607.
	WANG Liying, WANG Xinning. Multi-value voxel connected region construction based on 3D plane extraction for airborne LiDAR data[J]. Journal of Geo-information Science, 2021, 23(9): 1598-1607.
[27]	ROTTENSTEINER F, SOHN G, JUNG J, et al. The ISPRS benchmark on urban object classification and 3D building reconstruction[EB/OL]. [2023-09-10]. https://isprs-annals.copernicus.org/articles/I-3/293/2012/isprsannals-I-3-293-2012.pdf.
[28]	ZHAO Zongze, DUAN Yansong, ZHANG Yongjun, et al. Extracting buildings from and regularizing boundaries in airborne LiDAR data using connected operators[J]. International Journal of Remote Sensing, 2016, 37(4): 889-912.
[29]	ZAREA A, MOHAMMADZADEH A. A novel building and tree detection method from LiDAR data and aerial images[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2016, 9(5): 1864-1875.
[30]	NGUYEN T H, DANIEL S, GUÉRIOT D, et al. Super-resolution-based snake model—an unsupervised method for large-scale building extraction using airborne LiDAR data and optical image[J]. Remote Sensing, 2020, 12(11): 1702.
[31]	GILANI S A N, AWRANGJEB M, LU Guojun. An automatic building extraction and regularisation technique using LiDAR point cloud data and orthoimage[J]. Remote Sensing, 2016, 8(3): 258.

区域	空间邻域	一类误差/（%）	二类误差/（%）	总误差/（%）
Area1	6	84.87	0.91	43.12
	18	11.56	32.78	22.11
	26	6.52	5.75	6.14
	56	2.26	4.93	3.59
	80	4.85	9.65	7.23
	124	4.38	14.59	9.46
Area2	6	50.93	2.97	15.88
	18	18.25	2.21	6.53
	26	14.44	3.39	6.37
	56	3.90	0.65	1.53
	80	4.48	1.10	2.01
	124	3.82	1.05	1.80
Area3	6	77.50	1.45	32.98
	18	24.89	5.64	13.62
	26	16.09	3.12	8.50
	56	4.31	2.83	3.45
	80	4.92	8.34	6.92
	124	4.97	18.64	12.98

区域	完整率	正确率	质量
Area1	99.0	98.0	97.0
Area2	99.2	98.1	96.4
Area3	98.9	97.9	96.9

建筑物提取算法	Area1			Area2			Area3
建筑物提取算法	完整率	正确率	质量	完整率	正确率	质量	完整率	正确率	质量
文献[28]	89.4	94.5	84.9	91.0	95.0	86.8	87.7	94.5	83.4
文献[29]	91.4	94.3	86.6	86.5	93.6	81.7	88.3	99.0	87.5
文献[30]	90.4	94.2	85.7	93.5	94.8	88.9	91.0	93.0	85.2
文献[31]	85.4	86.4	75.4	88.8	84.5	76.4	89.9	84.7	77.4
文献[19]	90.6	94.5	86.0	93.8	98.7	92.7	85.5	94.9	85.7
本文方法	99.0	98.0	97.0	99.2	98.1	96.4	98.9	97.9	96.9