3D点云震害建筑物深度学习样本增强方法

  • role: First author第一作者
  • Affiliation:

    Laboratory of Earthquake Numerical Prediction and Risk Prediction, Institute of Earthquake Forecasting, CEA, Beijing 100036, China

  • Email:15738510390@163.com
  • Introduction:驿GISE-mail15738510390@163.com
CUI Yining1,  
  • role: Corresponding author通信作者
  • Affiliation:

    Laboratory of Earthquake Numerical Prediction and Risk Prediction, Institute of Earthquake Forecasting, CEA, Beijing 100036, China

  • Email:axdothy@163.com
  • Introduction:E-mailaxdothy@163.com
DOU Aixia1*,  
  • Affiliation:

    Beijing MAG Tianhong Science&Technology Development Co., Ltd, Beijing 100089, China

YANG Shenning2

résumé

Pour résoudre le problème de l'identification automatique des types de dommages aux bâtiments à partir des nuages de points LiDAR dans des scénarios complexes après un séisme, pour répondre aux exigences de rapidité et de précision dans les situations d'urgence, pour dire adieu à l'extraction manuelle traditionnelle des caractéristiques de dommages et exploiter pleinement les informations sur les dommages aux bâtiments dans les zones sinistrées, et pour réaliser ensuite l'identification automatique et intelligente des bâtiments. Cet article applique la méthode d'apprentissage profond des nuages de points 3D à l'identification des dommages aux bâtiments, a créé une base de données de nuages de points contenant des types de dommages tels que l'effondrement, l'effondrement partiel et le non-effondrement. Le réseau PointNet++ examine l'impact du volume et de l'équilibre des échantillons sur la précision de l'identification, et propose une méthode d'amélioration des échantillons de bâtiments endommagés, enrichissant les formes de points dans les échantillons de différentes classes. Sur la base des données LiDAR après le séisme de magnitude 7,0 en Haïti en 2010, la précision de la classification avant et après l'amélioration des échantillons dans le réseau PointNet++ a été comparée, les résultats ont montré une amélioration de la précision de la classification des effondrements et des effondrements partiels de 27% et 17% respectivement après l'amélioration des échantillons, et une augmentation de 15% de la précision moyenne de la classification et du coefficient Kappa. Les résultats des expériences montrent que le modèle d'apprentissage profond en trois dimensions des dommages aux bâtiments présente une bonne amélioration de la classification lorsque le volume et l'équilibre des échantillons sont suffisants pour toutes les classes d'échantillons.

mots-clés

remote sensing;classification and recognition;PointNet++;sample enhancement;LiDAR point cloud;seismic damage buildings

References

  1. 1.
    Deng F, Dou A X and Wang X Q. 2018. Fusion of aerial imagery and airborne LiDAR data for post-earthquake building point extraction. Journal of Remote Sensing, 22(S1): 224-232
  2. 2.
    Dou A X. 2018. Study on Earthquake Damage Identification Feature Parameters of Buildings Based on Airborne LiDAR Data. Beijing: Institute of Geology, Earthquake Administration
  3. 3.
    Gerke M and Xiao J. 2014. Fusion of airborne laserscanning point clouds and images for supervised and unsupervised scene classification. ISPRS Journal of Photogrammetry and Remote Sensing, 87: 78-92
  4. 4.
    Gong W P, Li Y F, Chen Z and Long C. 2013. A study of fast building extraction from airborne LiDAR point cloud data. Urban Geotechnical Investigation and Surveying, (3): 84-88
  5. 5.
    Hu B G. 2013. Object-Oriented Damage Building Extraction Based on Lidar and High Resolution Remote Sensing Imagery. Chengdu: Southwest Jiaotong University
  6. 6.
    Huang S S, Dou A X, Wang X Q and Yuan X X. 2016. Building damage feature analyses based on post-earthquake airborne LiDAR data. Acta Seismologica Sinica, 38(3): 467-476
  7. 7.
    Kalogerakis E, Averkiou M, Maji S and Chaudhuri. 2017. 3D shape segmentation with projective convolutional networks//Proceedings of 2017 Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu: IEEE
  8. 8.
    Klokov R and Lempitsky V. 2017. Escape from cells: deep Kd-networks for the recognition of 3D point cloud models//Proceedings of 2017 IEEE International Conference on Computer Vision (ICCV). Venice: IEEE. 863-872
  9. 9.
    Li Y Y, Bu R, Sun M C, Wu W, Di X H and Chen B Q. 2018. PointCNN: convolution on X-transformed points//Proceedings of the 32nd International Conference on Neural Information Processing Systems. Montreal: Curran Associates Inc
  10. 10.
    Lu Q, Chen C, Xie W J and Luo Y T. 2020. PointNGCNN: deep convolutional networks on 3D point clouds with neighborhood graph filters. Computers and Graphics, 86: 42-51
  11. 11.
    Ma W, Yue J P and Cao S. 2010. Building edge extraction from LIDAR data based on images segmentation. Geography and Geographic Information Science, 26(4): 57-59, 63
  12. 12.
    Maturana D and Scherer S. 2015. VoxNet: A 3D convolutional neural network for real-time object recognition//Proceedings of 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). Hamburg: IEEE
  13. 13.
    Qi C R, Su H, Mo K, Kaichun M and Guibas L J. 2017a. PointNet: deep learning on point sets for 3D classification and segmentation//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE
  14. 14.
    Qi C R, Yi L, Su H and Guibas L J. 2017b. PointNet++: deep hierarchical feature learning on point sets in a metric space//Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach: Curran Associates Inc
  15. 15.
    Riegler G, Ulusoy A O and Geiger A. 2017. OctNet: learning deep 3D representations at high resolutions//Proceedings of 2017 Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu: IEEE: 3577-3586
  16. 16.
    Sinha A, Bai J and Ramani K. 2016. Deep learning 3D shape surfaces using geometry images//Proceedings of the 14th European Conference on Computer Vision. Amsterdam: Springer: 223-240
  17. 17.
    Su H, Maji S, Kalogerakis E and Learned-Miller E. 2015. Multi-view convolutional neural networks for 3D shape recognition//Proceedings of 2015 IEEE International Conference on Computer Vision. Santiago: IEEE
  18. 18.
    Turker M and Koc-San D. 2015. Building extraction from high-resolution optical spaceborne images using the integration of support vector machine (SVM) classification, Hough transformation and perceptual grouping. International Journal of Applied Earth Observation and Geoinformation, 34: 58-69
  19. 19.
    Wang C F. 2013. Detection of Collapsed Building by Classifying Airborne Laser Scanner Data. Chengdu: Southwest Jiaotong University
  20. 20.
    Wang J X, Dou A X, Wang X Q, Huang S S and Zhang X H. 2017. The ground-objects classification based on post-earthquake airborne LiDAR Data. Technology for Earthquake Disaster Prevention, 12(3): 677-689
  21. 21.
    Wu Z R, Song S R, Khosla A, Yu F, Zhang L G, Tang X O and Xiao J X. 2015. 3D ShapeNets: a deep representation for volumetric shapes//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Boston: IEEE
  22. 22.
    Yu H Y, Cheng G, Zhang Y M and Lu X P. 2011. The detection of earthquake-caused collapsed building information from LiDAR data and aerophotograph. Remote Sensing for Land and Resources, (3): 77-81
  23. 23.
    Zhang W M, Qi J B, Wan P, Wang H T, Xie D H, Wang X Y and Yan G J. 2016. An easy-to-use airborne LiDAR data filtering method based on cloth simulation. Remote Sensing, 8(6): 501
  24. 24.
    Zhao H S, Jiang L, Fu C W and Jia J Y. 2019. PointWeb: enhancing local neighborhood features for point cloud processing//Proceedings of 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Long Beach: IEEE: 5565-5573

Lire l'article complet

The above content is generated by Large Model Translation. The translated content is for reference only. We do not assume any commercial or legal responsibilty for any consequences arising from the use of our website