地球信息科学学报 ›› 2023, Vol. 25 ›› Issue (5): 1075-1087.doi: 10.12082/dqxxkx.2023.220736

• 遥感科学与应用技术 • 上一篇    

基于特征增强核点卷积网络的多光谱LiDAR点云分类方法

陈科(), 管海燕(), 雷相达, 曹爽   

  1. 南京信息工程大学遥感与测绘学院,南京 210044
  • 收稿日期:2022-09-28 修回日期:2022-12-14 出版日期:2023-05-25 发布日期:2023-04-27
  • 通讯作者: *管海燕(1976— ),女,江苏南京人,博士,教授,主要从事遥感数据智能解译。E-mail: guanhy.nj@nuist.edu.cn
  • 作者简介:陈 科(1998— ),男,江苏南京人,硕士,主要从事三维点云数据处理。E-mail: 20201248038@nuist.edu.cn
  • 基金资助:
    国家自然科学基金项目(41971414)

A Multispectral LiDAR Point Cloud Classification Method based on Enhanced Features Kernel Point Convolutional Network

CHEN Ke(), GUAN Haiyan(), LEI Xiangda, CAO Shuang   

  1. School of Remote Sensing & Geomatics Engineering, Nanjing University of Information Science & Technology, Nanjing 210044, China
  • Received:2022-09-28 Revised:2022-12-14 Online:2023-05-25 Published:2023-04-27
  • Contact: GUAN Haiyan
  • Supported by:
    National Natural Science Foundation of China(41971414)

摘要:

多光谱LiDAR系统可同时提供目标地物的三维空间和光谱的信息,为地物识别、土地覆盖分类以及场景理解提供了便利。针对目前大规模多光谱LiDAR点云几何信息挖掘不充分与数据处理效率低问题,本文提出一种利用局部点云几何信息增强特征的端到端连续核点卷积网络—特征增强核点卷积网络的机载多光谱LiDAR点云分类方法。该网络是一个编解码结构,网络结构中结合随机采样与最远点采样快速处理大规模多光谱LiDAR点云,保证采样效率的同时减少随机采样导致的关键几何结构信息缺失。为提高多光谱LiDAR点云数据语义信息表达,设计了基于连续可变卷积的增强卷积模块,在聚合局部特征的同时,利用邻近点与中心点的位置关系增强赋予核点的局部特征;最后结合加权标签平滑损失与Lovasz-Softmax损失进一步提升多光谱LiDAR点云分类性能。通过对加拿大Optech 公司的Titan多光谱LiDAR点云数据集的实验表明,本文所提出的特征增强核点卷积网络的综合精度、macro-F1指数和mIoU值分别为96.80%、88.51%和83.42%,相较于同类型网络在多光谱LiDAR点云分类性能上具有一定优越性;使用格网采样与KD-Tree结合更好地保留原始点云的几何特征;在单批次65 536个点情况下,点云采样时间较同类多光谱LiDAR点云分类网络减少28261.79 ms,实现快速处理海量点云。实验结果证明了特征增强核点卷积网络在多光谱LiDAR点云分类任务上的潜力。

关键词: 多光谱LiDAR, 卷积神经网络, 核点卷积, 点云分类, 特征增强, 损失函数, 点云采样

Abstract:

The multispectral LiDAR system can simultaneously provide the 3D space and spectral information of the target ground object, which is convenient for ground object recognition, land cover/use classification, and scene understanding. However, most multispectral LiDAR point cloud classification methods cannot fully mine the geometric information of point clouds and achieve poor performance in fine-scale classification. To overcome this limitation, this paper presents a continuous kernel point convolutional network which uses local point cloud geometric information to enhance features. Firstly, the network combines a random sampling with a farthest point sampling to quickly process large-scale multispectral LiDAR point clouds. Then, an enhanced convolution module based on continuous variable convolution is designed to improve the semantic information expression of multispectral LiDAR point cloud data. In order to address the problem that kernel point convolution simply using the distance relationship between the geometric space and feature space of neighboring points and centroids is insufficient to express the local information as a complementary feature of the kernel point convolution network, the local features given to the kernel points are enhanced by using the position relationship between neighboring points and centroids while aggregating the local features to provide richer semantic information for the multispectral LiDAR point cloud classification network. Finally, the weighted label smoothing loss and the Lovasz-Softmax loss are combined to further improve the classification performance. The results on the Titan multispectral LiDAR dataset show that the proposed network achieves an overall accuracy of 96.80%, a macro-F1 index of 88.51%, and a mIoU value of 83.42%, which is superior to the state-of-the-art (SOTA) multispectral LiDAR data networks. The proposed model uses the combination of grid sampling and KD-Tree to better preserve the geometric features of the original point cloud. In the case of a single batch of 65,536 points, the point cloud sampling time is reduced by 28 261.79 ms compared with similar multispectral LiDAR point cloud classification networks. This Study demonstrates the potential of enhanced feature kernel points convolutional network for multispectral LiDAR point cloud classification tasks.

Key words: multispectral LiDAR, kernel point convolution, convolutional neural network, point cloud classification, feature enhancement, loss function, point cloud sampling