Automatic classification of pole-like objects in road scene by back propagation neural network

doi:10.13474/j.cnki.11-2246.2020.0120

Abstract

Abstract: Aiming at the problems of low accuracy and low automation of pole-like objects classification in vehicle laser scanning data, a classification method based on BP neural network is proposed. Firstly, according to the point cloud characteristics of the pole-like object, ten eigenvalues are selected to obtain the feature vectors of the pole-like object clustering unit, and then eigenmatrix is constructed. Secondly, the BP neural network model is trained by using the sample set and the classification model is saved. Finally, the BP neural network classification model is used to classify the pole-like object in the test area. The experiment showed that the classification accuracy of the method for pole-like objects could reach 95.34%, which also verifies the effectiveness of the method.

Key words: pole-like objects, neural network, automatic classification, eigenvalues, classification model

CLC Number:

LI Pengpeng, LI Yongqiang, ZHAO Shangbin, FAN Huilong. Automatic classification of pole-like objects in road scene by back propagation neural network[J]. Bulletin of Surveying and Mapping, 2020, 0(4): 101-105,120.

References

[1] 李永强, 董亚涵, 张西童, 等. 车载LiDAR点云路灯提取方法[J]. 测绘学报, 2018,47(2):247-259.
[2] 谭奔,钟若飞,李芹.车载激光扫描数据的地物分类方法[J].遥感学报,2012,16(1):50-66.
[3] 冯从义,岑敏仪,张同刚.基于知识的车载LiDAR地物自动分类[J].计算机工程与应用,2016,52(5):122-126.
[4] 王鹏,刘如飞,马新江,等.一种车载激光点云中杆目标自动提取方法[EB/OL].(2017-12-21)[2018-09-07].https://doi.org/10.13203/j.whugis20170421.
[5] 马东岭,王晓坤,李广云.一种基于高度差异的点云数据分类方法[J].测绘通报,2018(6):46-49.
[6] 张西童,刘会云,李永强,等.车载LiDAR场景中路灯的提取和识别[J].测绘工程,2016,25(9):50-54.
[7] YU Y T, LI J, GUAN H Y, et al. Semiautomated extraction of street light poles from mobile LiDAR point-clouds[J].IEEE Transactions on Geoscience and Remote Sensing, 2014,53(3):1374-1386.
[8] YUT S, WANG R S. Scene parsing using graph matching on street-view data[J]. Computer Vision and Image Understanding, 2016,145:70-80.
[9] 乔纪纲,陈明辉,艾彬,等.SVM用于LiDAR数据的地物分类[J].测绘通报,2013(7):35-42.
[10] YI L, SU H, GUO X, et al. SyncSpecCNN: synchronized spectral CNN for 3D shape segmentation[C]//Computer Vision and Pattern Recognition. [S.l.]: IEEE, 2017.
[11] 赵中阳,程英蕾,释小松,等.基于多尺度特征和PointNet的LiDAR点云地物分类方法[J].激光与光电子学进展,2019,56(5):251-258.
[12] 李海亭,王厚之,李艳红,等.支持向量机的车载雷达点云目标识别[J].测绘科学,2016,41(5):45-49.
[13] LEHTOMAKI M, JAAKKOLA A, HYYPPA J, et al. Detection of vertical pole-like objects in a road environment using vehicle-based laser scanning data [J]. Remote Sensing,2010,2(3):641-664.
[14] WU F, WEN C L, GUO Y L, et al. Rapid localization and extraction of street light poles in mobile LiDAR point clouds: a supervoxel-based approach[J]. IEEE Transactions on Intelligent Transportation Systems, 2017,18(2):292-305.
[15] GUAN H Y, YU Y T, LI J, et al. Pole-like road object detection in mobile LiDAR data via supervoxel and bag-of-contextual-visual-words representation[J]. IEEE Geoscience and Remote Sensing Letters, 2016,13(4):520-524.
[16] 李艳红.车载移动测量系统数据配准与分类识别关键技术研究[D].武汉:武汉大学,2014.
[17] RUMELHART D E, HINTON G E, WILLIAMS R J. Learning representations by back-propagating errors[J]. Nature, 1986,323(6088):533-536.