With the development of unmanned aerial vehicle (UAV) LiDAR technology, large-scale high-precision point cloud data is gradually playing an important role in the classification and extraction of terrain and features. This paper constructs a point cloud classification model based on an improved convolutional neural network (CNN) for the surface unmanned aerial vehicle (UAV) LiDAR point cloud data in a certain area of Sichuan Province, with ground points, building points, and vegetation points as categories. Firstly, the categories of the original point cloud are manually labeled to create the training set and test set, and the geometric and intensity features of each point are extracted. Then, a multi-layer CNN model is constructed using PyTorch to conduct end-to-end classification training and model testing on the point cloud. The experimental results show that the overall classification accuracy (OA) can reach 93.6599%, and the classification effect on ground points and vegetation points is especially good.
Cite this paper
Guo, Y. (2025). Research on LiDAR Point Cloud Terrain Classification Based on Convolutional Neural Network. Open Access Library Journal, 12, e13925. doi: http://dx.doi.org/10.4236/oalib.1113925.
Diab, A., Kashef, R. and Shaker, A. (2022) Deep Learning for Lidar Point Cloud Classification in Remote Sensing. Sensors, 22, Article 7868. https://doi.org/10.3390/s22207868
Alfio, V.S., Pepe, M. and Costantino, D. (2024) The Use of Random Forest for the Classification of Point Cloud in Urban Scene. International Journal of Engineering Trends and Tech-nology, 72, 1-9. https://doi.org/10.14445/22315381/ijett-v72i3p101
Su, H., Maji, S., Kalogerakis, E. and Learned-Miller, E. (2015). Multi-View Convolutional Neural Networks for 3D Shape Recognition. 2015 IEEE International Conference on Computer Vision (ICCV), Santiago, 7-13 December 2015, 945-953. https://doi.org/10.1109/iccv.2015.114
Qi, C.R., Su, H., Mo, K. and Guibas, L.J. (2017) PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, 21-26 July 2017, 77-85.
Qi, C.R., Yi, L. and Su, H. (2017) PointNet : Deep Hierarchical Fea-ture Learning on Point Sets in a Metric Space. Proceedings of the 31st International Conference on Neural Information Pro-cessing Systems, Long Beach, 4-9 December 2017, 5105-5114.
Li, Y., Ma, L., Zhong, Z., Liu, F., Chapman, M.A., Cao, D., et al. (2021) Deep Learning for Lidar Point Clouds in Autonomous Driving: A Review. IEEE Transactions on Neural Networks and Learning Systems, 32, 3412-3432. https://doi.org/10.1109/tnnls.2020.3015992
Bello, S.A., Yu, S., Wang, C., Adam, J.M. and Li, J. (2020) Review: Deep Learning on 3D Point Clouds. Remote Sensing, 12, Article 1729. https://doi.org/10.3390/rs12111729
Song, W., Zhang, L., Tian, Y., Fong, S., Liu, J. and Gozho, A. (2020) CNN-Based 3D Object Classification Using Hough Space of Lidar Point Clouds. Human-Centric Computing and Information Sciences, 10, Article No. 19. https://doi.org/10.1186/s13673-020-00228-8
Li, X., Wang, M., Wen, C., Wang, L., Zhou, N. and Fang, Y. (2019) Density-Aware Convolutional Networks with Context Encoding for Airborne LiDAR Point Cloud Classification. arXiv: 1910.05909.
Matti, D., Ekenel, H.K. and Thiran, J. (2017) Combining Lidar Space Clustering and Convolutional Neural Networks for Pedestrian Detection. 2017 14th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS), Lecce, 29 August-1 September 2017, 1-6. https://doi.org/10.1109/avss.2017.8078512
Gamal, A., Wibisono, A., Wicaksono, S.B., Abyan, M.A., Hamid, N., Wisesa, H.A., et al. (2020) Automatic LIDAR Building Segmenta-tion Based on DGCNN and Euclidean Clustering. Journal of Big Data, 7, Article No. 102. https://doi.org/10.1186/s40537-020-00374-x