| 基于深度残差网络的机载LiDAR点云分类 |
| |
| 引用本文： | 赵传,郭海涛,卢俊,余东行,张保明.基于深度残差网络的机载LiDAR点云分类[J].测绘学报,2020,49(2):202-213. |
| |
| 作者姓名： | 赵传 郭海涛 卢俊 余东行 张保明 |
| |
| 作者单位： | 信息工程大学地理空间信息学院, 河南 郑州 450001 |
| |
| 基金项目： | 国家自然科学基金（41601507） |
| |
| 摘 要： | 机载LiDAR点云的分类是利用其进行城市场景三维重建的关键步骤之一。为充分利用现有的图像领域性能较好的深度学习网络模型,提高点云分类精度,并降低训练时间和对训练样本数量的要求,本文提出一种基于深度残差网络的机载LiDAR点云分类方法。首先提取归一化高程、表面变化率、强度和归一化植被指数4种具有较高区分度的点云低层次特征;然后通过设置不同的邻域大小和视角,利用所提出的点云特征图生成策略,得到多尺度和多视角点云特征图;再将点云特征图输入到预训练的深度残差网络,提取多尺度和多视角深层次特征;最后构建并训练神经网络分类器,利用训练的模型对待分类点云进行预测,经后处理得到分类结果。利用ISPRS三维语义标记竞赛的公开标准数据集进行试验,结果表明,本文方法可有效区分建筑物、地面、车辆等8类地物,分类结果的总体精度为87.1%,可为城市场景三维重建提供可靠的信息。
|
| 关 键 词： | 点云分类 深层次特征 多尺度和多视角 迁移学习 深度残差网络 机载LIDAR |
| 收稿时间： | 2019-01-02 |
| 修稿时间： | 2019-08-29 |
Airborne LiDAR point cloud classification based on deep residual network |
| |
| ZHAO Chuan,GUO Haitao,LU Jun,YU Donghang,ZHANG Baoming.Airborne LiDAR point cloud classification based on deep residual network[J].Acta Geodaetica et Cartographica Sinica,2020,49(2):202-213. |
| |
| Authors: | ZHAO Chuan GUO Haitao LU Jun YU Donghang ZHANG Baoming |
| |
| Institution: | Institute of Geospatial Information, Information Engineering University, Zhengzhou 450001, China |
| |
| Abstract: | Airborne LiDAR point cloud classification is one of the key steps for three-dimensional reconstruction of urban scenes. To leverage the existing high-performing deep learning network model in image field of image processing, improve classification accuracy and reduce training time and demand for training samples simultaneously, an airborne LiDAR point cloud classification method based on deep residual network is proposed in this paper. Firstly, high discriminative low-level features, i.e. normalized height, point cloud normal vector, intensity and normalized differential vegetation index, are extracted. Secondly, by setting different neighborhood sizes and perspectives, multi-scale and multi-view point cloud feature images are generated via using the proposed point cloud feature image generation strategy. Then, point cloud feature images are input into the pre-trained deep residual network to extract multi-scale and multi-view deep features. Finally, a neural network classifier is constructed and trained, point cloud classification results are obtained by utilizing the trained classifier and postprocessing. Benchmark datasets of ISPRS 3D Semantic Labeling Contest are used, the experimental results show that the proposed method can effectively distinguish 8 types ground objects such as buildings, ground and vehicles etc., and the overall accuracy of the classification result is 87.1%, which can provide reliable information for three-dimensional reconstruction of urban scenes. |
| |
| Keywords: | point cloud classification deep feature multi-scale and multi-view transfer learning deep residual network airborne LiDAR |
| 本文献已被 CNKI 维普 等数据库收录！ |
| 点击此处可从《测绘学报》浏览原始摘要信息 |
| 点击此处可从《测绘学报》下载免费的PDF全文 |
|
