Automatic building extraction from oblique photogrammetry data by combining height and spectral information

doi:10.13474/j.cnki.11-2246.2022.0035

Abstract

Abstract: Building is a major component of 3D model and its vectorization mainly relies on manual delineation. Some studies have investigated the building extraction using deep learning methods, however, such methods require large volume of labeled training samples. To solve these problems, this paper proposes a method for automatic building extraction by combining height and spectral information of oblique photogrammetry data in typical areas of Tianjin. The method employes height threshold segmentation, vegetation elimination, and morphological post-processing to gradually refine the building extraction results, which achieves the overall accuracy of 94%. The extracted building footprints are vectorized and regularized, and then the individual building can be queried in geographic information system which can promote the application of 3D real-scene model.

Key words: building extraction, oblique photogrammetry, 3D real-scene model, automation, vectorization

CLC Number:

WANG Zhen, ZHANG Tao, DING Lele, PAN Yuming, SHI Furong. Automatic building extraction from oblique photogrammetry data by combining height and spectral information[J]. Bulletin of Surveying and Mapping, 2022, 0(2): 10-15.

References

[1] 李鹏鹏,孙雅庚.实景三维与多源数据融合技术的研究和应用[J].测绘通报, 2019(S1):133-136.
[2] SPORTOUCHE H, TUPIN F, DENISE L. Extraction and three-dimensional reconstruction of isolated buildings in urban scenes from high-resolution optical and SAR spaceborne images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2011, 49(10):3932-3946.
[3] 王树臻,郑国强,王光生,等.多源点云数据融合的建筑物精细化建模[J].测绘通报, 2020(8):28-32, 38.
[4] 郭功举,林木棵,刘一宁.地理实体数据库研究与实践[J].测绘地理信息, 2021, 46(6):127-131.
[5] 蔡香玉.基于无人机倾斜摄影场景建模的建筑物单体化方法[D].南京:南京师范大学, 2018.
[6] 郭如宝,郝利娟,刘旭东,等.基于倾斜摄影的实景三维单体化模型构建方法[J].测绘通报, 2021(1):103-107.
[7] HUANG Xin, ZHANG Liangpei. Morphological building/shadow index for building extraction from high-resolution imagery over urban areas[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2012, 5(1):161-172.
[8] 王珍,张涛,丁乐乐,等.高分辨率影像深度学习概率决策融合建筑物提取[J].测绘科学, 2021, 46(6):93-101.
[9] 张涛,丁乐乐,史芙蓉.高分辨率遥感影像城中村提取的景观语义指数方法[J].测绘学报, 2021, 50(1):97-104.
[10] NGO T T, MAZET V, COLLET C, et al. Shape-based building detection in visible band images using shadow information[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2017, 10(3):920-932.
[11] ZHANG Liangpei, ZHANG Lefei, DU Bo. Deep learning for remote sensing data:a technical tutorial on the state of the art[J]. IEEE Geoscience and Remote Sensing Magazine, 2016, 4(2):22-40.
[12] BOONPOOK W, TAN Yumin, XU Bo. Deep learning-based multi-feature semantic segmentation in building extraction from images of UAV photogrammetry[J]. International Journal of Remote Sensing, 2021, 42(1):1-19.
[13] 张涛,丁乐乐,史芙蓉.融合类别语义特征的卷积神经网络建筑物提取[J].遥感信息, 2021, 36(5):49-55.
[14] 戴玉成.基于UAV倾斜影像匹配点云的城市建筑物信息提取方法研究[D].北京:中国科学院大学, 2017.
[15] LIU Chun, HUANG Xin, WEN Dawei, et al. Assessing the quality of building height extraction from ZiYuan-3 multi-view imagery[J]. Remote Sensing Letters, 2017, 8(9):907-916.
[16] ZHAO Zongze, WANG Hongtao, WANG Cheng, et al. Fusing LiDAR data and aerial imagery for building detection using a vegetation-mask-based connected filter[J]. IEEE Geoscience and Remote Sensing Letters, 2019, 16(8):1299-1303.
[17] MEYER G E, NETO J C. Verification of color vegetation indices for automated crop imaging applications[J]. Computers and Electronics in Agriculture, 2008, 63(2):282-293.
[18] GAO Xianjun, WANG Mingwei, YANG Yuanwei, et al. Building extraction from RGB VHR images using shifted shadow algorithm[J]. IEEE Access, 2018, 6:22034-22045.
[19] LI Weijia, HE Conghui, FANG Jiarui, et al. Semantic segmentation-based building footprint extraction using very high-resolution satellite images and multi-source GIS data[J]. Remote Sensing, 2019, 11(4):403.
[20] GRIBOV A. Searching for a compressed polyline with a minimum number of vertices[C]//Proceedings of 2017 IAPR International Conference on Document Analysis and Recognition (ICDAR).Kyoto:IEEE, 2017:13-14.