Automatic division and regularization of indoor 3D point cloud space

doi:10.13474/j.cnki.11-2246.2022.0238

Abstract

Abstract: Accurate spatial division is an important basis for realizing indoor semantic modeling and topology reconstruction.As a commonly used indoor space data carrier, 3D point cloud is of great significance to extract and normalize semantic information in indoor space based on 3D point cloud.This paper proposes an indoor scene segmentation based on the morphological segmentation method, and combines the vector regularization method.Firstly, spatial segmentation elements are extracted based on regional growth algorithm and linear fitting method, generate binary image through plane projection, and then use distance transformation and watershed algorithm to complete spatial segmentation. Secondly, linear fit of indoor spatial mesh.This paper verifies 4 sets of actual scene data, including 3 sets of ISPRS data sets and 1 set of actual scene collection data. The test results show that the indoor space segmentation and regularization method proposed in this paper can accurately and quickly complete the extraction of indoor space elements.

Key words: 3D point cloud, spatial section, watershed, regularization

CLC Number:

P237

WANG Jingchun, TANG Shengjun, WANG Weixi, LI Xiaoming, LI You, XIE Linfu, ZHU Jiasong. Automatic division and regularization of indoor 3D point cloud space[J]. Bulletin of Surveying and Mapping, 2022, 0(8): 93-97,138.

References

[1] 刘先林.为社会进步服务的测绘高新技术[J].测绘科学,2019,44(6):1-15.
[2] 赵江洪,董岩,危双丰,等.室内导航路网提取研究进展[J].测绘科学,2020,45(12):45-54.
[3] FABRIZI E,SAFFIFIOTTI A. Augmenting topology-based maps with geometric information[J]. Robotics and Autonomous Systems,2002,40(2): 91-97.
[4] JAEHOON J,CYRILL S,CHANGJAE K. Automatic room segmentation of 3D laser data using morphological processing[J]. ISPRS international Journal of Geo-Information,2017,6: 206.
[5] 黄帅.基于点云的室内语义模型自动构建技术[D]. 北京: 北京建筑大学,2020.
[6] XIE L,WANG R. Automatic indoor building reconstruction from mobile laser scanning data[J]. International Archives of the Photogrammetry,Remote Sensing and Spatial Information Sciences,2017,XLII-2/W7: 417-422.
[7] KHOSHELHAM K,DÍAZ VILARIÑO L,PETER M,et al. The ISPRS benchmark on indoor modeling[J]. International Archives of the Photogrammetry,Remote Sensing and Spatial Information Sciences,2017,XLII-2/W7:367-372.
[8] 韩李涛,周丽娟,龚城,等.顾及步行习惯的室内导航网络及其生成算法[J].测绘学报,2022,51(5):729-738.
[9] 宋锦阳,钱建国,汤圣君,等.三维点云室内平面要素提取与优化[J].测绘通报,2021(9):15-20.
[10] 孔繁圣,李子印,古亭,等.基于多层次距离变换的蚕茧分割计数方法[J].蚕业科学,2020,46(2):190-196.
[11] 张霞,张春亢,李红梅,等.结合笔画宽度变换与均值漂移的遥感影像道路提取 [J].测绘科学技术学报,2019,36(3):287-292.
[12] 桑艳艳,李昕.基于改进分水岭算法的菌落图像分割[J]. 电子测量技术,2019,42(6): 87-93.
[13] 崔扬.移动LiDAR点云室内三维结构化重建方法和关键技术研究[J].测绘学报,2021,50(7):990.
[14] 李灿.面向停车诱导的室内地图模型构建与应用[D].武汉:武汉大学,2017.
[15] BORMANN R,JORDAN F,LI W,et al. Room segmentation: survey,implementation,and analysis[C]//Proceedings of 2016 IEEE International Conference on Robotics and Automation.Stockholm: IEEE,2016:1019-1022.
[16] AMBRUS R,CLAICI S,WENDT A. Automatic room segmentation from unstructured 3D data of indoor environments[J]. IEEE Robotics and Automation Letters,2017,2(2): 749-756.
[17] MACHER H,LANDES T,GRUSSENMEYER P. From point clouds to building information models: 3D semi-automatic reconstruction of indoors of existing buildings[J]. Applied Sciences,2017,7(10): 1030.
[18] LI L,SU F,YANG F,et al. Reconstruction of three-dimensional (3D) indoor interiors with multiple stories via comprehensive segmentation[J]. Remote Sensing,2018,10(8): 1281.
[19] YANG F,LI L,SU F,et al. Semantic decomposition and recognition of indoor spaces with structural constraints for 3D indoor modelling[J].Automation in Construction,2019,106(3):102913.