利用迁移学习的机载激光雷达点云分类

赵传; 余东行; 郭海涛; 卢俊

doi:10.3788/OPE.20192707.1601

您当前的位置：

首页 >

文章列表页 >

利用迁移学习的机载激光雷达点云分类

信息科学 | 更新时间：2020-08-13

- 利用迁移学习的机载激光雷达点云分类
- Airborne LiDAR point cloud classification using transfer learning
- 光学精密工程 2019年27卷第7期页码：1601-1612
- 作者机构：
  
  战略支援部队信息工程大学地理空间信息学院，河南郑州 450001
- 作者简介：
  
  [ "赵传 (1991-)，男，湖南湘潭人，博士研究生，2014年、2017年于解放军信息工程大学分别获得学士、硕士学位，主要从事机载LiDAR点云数据处理与建筑物三维模型重建，数字摄影测量的研究。E-mail:zc_mail163@163.com" ]
- 基金信息：
  
  国家自然科学基金资助项目(41601507);地理信息工程国家重点实验室开放基金资助项目(SKLGIE2017-M-3-2)
- DOI：10.3788/OPE.20192707.1601
  中图分类号： P237
- 收稿日期：2018-10-16，
  
  录用日期：2018-12-14，
  
  纸质出版日期：2019-07-15
- 稿件说明：
移动端阅览
赵传, 余东行, 郭海涛, 等. 利用迁移学习的机载激光雷达点云分类[J]. 光学精密工程, 2019,27(7):1601-1612.

Chuan ZHAO, Dong-hang YU, Hai-tao GUO, et al. Airborne LiDAR point cloud classification using transfer learning[J]. Optics and precision engineering, 2019, 27(7): 1601-1612.
赵传, 余东行, 郭海涛, 等. 利用迁移学习的机载激光雷达点云分类[J]. 光学精密工程, 2019,27(7):1601-1612. DOI： 10.3788/OPE.20192707.1601.

Chuan ZHAO, Dong-hang YU, Hai-tao GUO, et al. Airborne LiDAR point cloud classification using transfer learning[J]. Optics and precision engineering, 2019, 27(7): 1601-1612. DOI： 10.3788/OPE.20192707.1601.

摘要

为解决现有机载激光雷达点云分类方法存在难以在获得较高精度的点云分类结果的同时降低分类过程所需时间等问题，提出了一种利用迁移学习的机载激光雷达点云分类方法。首先，计算点云的归一化高程、强度和法向量三个特征，通过设置不同邻域大小，利用所提出的点云特征图生成策略生成多尺度点云特征图；然后，利用预训练的深度残差网络从每个点的多尺度点云特征图提取其多尺度深度特征；最后，为了实现快速地训练，构建仅包含两层全连接神经网络模型，再利用训练好的模型对点云进行分类。两组ISPRS提供的标准点云数据集的试验结果表明：提出的方法所需训练时间少，分类结果的整体精度为89.6%，较ISPRS官网上所报道的最佳点云方法分类精度高4.4%。分类结果可为机载激光雷达点云的后续处理与应用提供可靠的信息。

Abstract

In order to overcome the problem that existing airborne methods for LiDAR point cloud classification have difficulties in obtaining high classification accuracy and reducing processing time simultaneously

a method using transfer learning for classifying airborne LiDAR point clouds was proposed. First

normalized height

intensity

and normal vector were calculated for each LiDAR point

by setting different sizes of neighborhood

and multi-scale point cloud feature images were generated by utilizing the proposed feature image generation strategy. Subsequently

a pre-trained deep residual network was employed to extract multi-scale deep features from the generated multi-scale feature images. Finally

a neural network model containing only two fully connected layers was constructed to achieve efficient training

and each point cloud was classified by the trained neural network model. Experimental results of two ISPRS benchmark airborne LiDAR point cloud sets demonstrat that the proposed method requires less training time

and the overall classification accuracy obtained by the method is 89.6%

which is 4.4% higher than the best classification result reported on the ISPRS official website. The classification result can provide reliable information for further processing and application of point cloud.

关键词

Keywords

references

WEINMANN M, JUTZI B, HINZ S, et al .. Semantic point cloud interpretation based on optimal neighborhoods, relevant features and efficient classifiers[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2015, 105: 286-304.

DU S J, ZHANG Y S, ZOU Z R, et al .. Automatic building extraction from LiDAR data fusion of point and grid-based features[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2017, 130: 294-307.

杨俊涛, 康志忠.多尺度特征和马尔可夫随机场模型的电力线场景点云分类法[J].测绘学报, 2018, 47(2): 188-197.

YANG J T, KANG ZH ZH. Multi-scale features and Markov random field model for powerline scene classification[J]. Acta Geodaetica Et Cartographica Sinica, 2018, 47(2): 188-197. (in Chinese)

张继贤, 段敏燕, 林祥国, 等.激光雷达点云电力线三维重建模型的对比与分析[J].武汉大学学报(信息科学版), 2017, 42(11): 1565-1572.

ZHANG J X, DUAN M Y, LIN X G, et al .. Comparison and analysis of models for 3D power line reconstruction using LiDAR point cloud[J]. Geomatics and Information Science of Wuhan University , 2017, 42(11): 1565-1572. (in Chinese)

张齐勇, 岑敏仪, 周国清, 等.城区LiDAR点云数据的树木提取[J].测绘学报, 2009, 38(4): 330-335.

ZHANG Q Y, CEN M Y, ZHOU G Q, et al .. Extracting trees from LiDAR data in urban region[J]. Acta Geodaetica Et Cartographica Sinica , 2009, 38(4): 330-335. (in Chinese)

BOYKO A, FUNKHOUSER T. Extracting roads from dense point clouds in large scale urban environment[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2011, 66(6): S2-S12.

VO A V, TRUONG-HONG L, LAEFER D F, et al .. Octree-based region growing for point cloud segmentation[J]. ISPRS Journal of Photogrammetry and Remote Sensing , 2015, 104: 88-100.

李明磊, 李广云, 王力, 等.采用八叉树体素生长的点云平面提取[J].光学精密工程, 2018, 26(1): 172-183.

LI M L, LI G Y, WANG L, et al .. Planar feature extraction from unorganized point clouds using octree voxel-based region growing[J]. Opt. Precision Eng ., 2018, 26(1): 172-183. (in Chinese)

王晓辉, 吴禄慎, 陈华伟, 等.应用改进的粒子群优化模糊聚类实现点云数据的区域分割[J].光学精密工程, 2017, 25(4): 563-573.

WANG X H, WU L SH, CHEN H W, et al .. Region segmentation of point cloud data based on improved particle swarm optimization fuzzy clustering[J]. Opt. Precision Eng ., 2017, 25(4): 563-573. (in Chinese)

ZHU Q, LI Y, HU H, et al .. Robust point cloud classification based on multi-level semantic relationships for urban scenes[J]. ISPRS Journal of Photogrammetry and Remote Sensing , 2017, 129: 86-102.

高仁强, 张显峰, 孙敏, 等.融合无人机LiDAR和高分辨率光学影像的点云分类方法[J].南京信息工程大学学报(自然科学版), 2018, 10(1): 102-112.

GAO R Q, ZHANG X F, SUN M, et al .. Fusion of high-resolution optical image and unmanned aerial vehicle LiDAR for 3D point cloud classification[J]. Journal of Nanjing University of Information Science & Technology(Natural Science Edition), 2018, 10(1): 102-112. (in Chinese)

潘锁艳, 管海燕.机载多光谱LiDAR数据的地物分类方法[J].测绘学报, 2018, 47(2): 198-207.

PAN S Y, GUAN H Y. Object classification using airborne multispectral LiDAR data[J]. Acta Geodaetica Et Cartographica Sinica , 2018, 47(2): 198-207. (in Chinese)

刘志青, 李鹏程, 陈小卫, 等.基于信息向量机的机载激光雷达点云数据分类[J].光学精密工程, 2016, 24(1): 210-219.

LIU ZH Q, LI P CH, CHEN X W, et al .. Classification of airborne LiDAR point cloud data based on information vector machine[J]. Opt. Precision Eng ., 2016, 24(1): 210-219. (in Chinese)

朱江涛, 黄睿.基于AdaBoost的高光谱与LiDAR数据特征选择与分类[J].遥感信息, 2014, 29(6):68-72.

ZHU J T, HUANG R. Feature selection and classification of hyperspectral data and LiDAR data based on AdaBoost[J]. Remote Sensing Information , 2014, 29(6):68-72. (in Chinese)

NIEMEYER J, WEGNER J D, MALLET C, et al .. Conditional random fields for urban scene classification with full waveform LiDAR data[C]. ISPRS Conference on Photogrammetric Image Analysis. Springer-Verlag, 2011: 233-244.

GUO B, HUANG X, ZHANG F, et al .. Classification of airborne laser scanning data using JointBoost[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2015, 100: 71-83.

NIEMEYER J, ROTTENSTEINER F, SOERGEL U. Contextual classification of LiDAR data and building object detection in urban areas[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2014, 87(1):152-165.

YANG Z S, JIANG W S, XU B, et al .. A convolutional neural network-based 3D semantic labeling method for ALS point clouds[J]. Remote Sensing, 2017, 9(9): 936.

ZHAO R, PANG M, WANG J. Classifying airborne LiDAR point clouds via deep features learned by a multi-scale convolutional neural network[J] . International Journal of Geographical Information Science , 2018(3):1-20.

YOUSEFHUSSIEN M, KELBE D J, IENTILUCCI E J, et al .. A multi-scale fully convolutional network for semantic labeling of 3D point clouds[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2018, 143: 191-204.

WANG J, LUO C, HUANG H Q, et al .. Transferring pre-trained deep CNNs for remote scene classification with general features learned from linear PCA network[J]. Remote Sensing, 2017, 9(3): 225.

RUSSAKOVSKY O, DENG J, SU H, et al .. ImageNet large scale visual recognition challenge[J]. International Journal of Computer Vision, 2015, 115(3): 211-252.

ZHANG W M, QI J B, WAN P, et al .. An easy-to-use airborne LiDAR data filtering method based on cloth simulation[J]. Remote Sensing, 2016, 8(6): 501.

BOULCH A, MARLET R. Fast and robust normal estimation for point clouds with sharp features[C]. Computer graphics forum. Blackwell Publishing Ltd , 2012, 31(5): 1765-1774.

TAN C Q, SUN F C, KONG T, et al .. A Survey on Deep Transfer Learning [M] Artificial Neural Networks and Machine Learning-ICANN 2018. Cham: Springer International Publishing, 2018: 270-279.

KORNBLITH S, SHLENS J, LE Q V. Do better ImageNet models transfer better?[J]. arXiv preprint arXiv : 1805. 08974, 2018.

HE K M, ZHANG X Y, REN S Q, et al .. Deep residual learning for image recognition[C]2016 IEEE Conference on Computer Vision and Pattern Recognition ( CVPR ), June 27-30, 2016. Las Vegas , NV , USA . New York , USA : IEEE , 2016.

CHARLES R Q, HAO S, MO K C, et al .. PointNet: deep learning on point sets for 3D classification and segmentation[C]2017 IEEE Conference on Computer Vision and Pattern Recognition ( CVPR ), July 21-26, 2017. Honolulu , HI . New York , USA : IEEE , 2017, 1(2): 4.

浏览量

121

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于ACGAN和迁移学习的骨显像分类方法

融合注意力的ConvNeXt视网膜病变自动分级

基于标签预留Softmax算法的机器视觉检测鉴别语义分割迁移学习技术

融合多尺度图像的密集神经网络肺部肿瘤识别算法

一种跨公司航天软件缺陷预测方法