当前位置: X-MOL 学术Int. J. Geograph. Inform. Sci. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Using multi-scale and hierarchical deep convolutional features for 3D semantic classification of TLS point clouds
International Journal of Geographical Information Science ( IF 5.7 ) Pub Date : 2018-12-10 , DOI: 10.1080/13658816.2018.1552790
Zhou Guo 1 , Chen-Chieh Feng 1
Affiliation  

ABSTRACT Point cloud classification, which provides meaningful semantic labels to the points in a point cloud, is essential for generating three-dimensional (3D) models. Its automation, however, remains challenging due to varying point densities and irregular point distributions. Adapting existing deep-learning approaches for two-dimensional (2D) image classification to point cloud classification is inefficient and results in the loss of information valuable for point cloud classification. In this article, a new approach that classifies point cloud directly in 3D is proposed. The approach uses multi-scale features generated by deep learning. It comprises three steps: (1) extract single-scale deep features using 3D convolutional neural network (CNN); (2) subsample the input point cloud at multiple scales, with the point cloud at each scale being an input to the 3D CNN, and combine deep features at multiple scales to form multi-scale and hierarchical features; and (3) retrieve the probabilities that each point belongs to the intended semantic category using a softmax regression classifier. The proposed approach was tested against two publicly available point cloud datasets to demonstrate its performance and compared to the results produced by other existing approaches. The experiment results achieved 96.89% overall accuracy on the Oakland dataset and 91.89% overall accuracy on the Europe dataset, which are the highest among the considered methods.

中文翻译:

使用多尺度和分层深度卷积特征进行 TLS 点云的 3D 语义分类

摘要 点云分类为点云中的点提供有意义的语义标签,对于生成三维 (3D) 模型至关重要。然而,由于不同的点密度和不规则的点分布,其自动化仍然具有挑战性。将现有的二维 (2D) 图像分类深度学习方法应用于点云分类是低效的,并且会导致对点云分类有价值的信息丢失。在本文中,提出了一种直接在 3D 中对点云进行分类的新方法。该方法使用深度学习生成的多尺度特征。它包括三个步骤:(1)使用3D卷积神经网络(CNN)提取单尺度深度特征;(2) 在多个尺度下对输入点云进行子采样,以每个尺度的点云作为3D CNN的输入,将多个尺度的深层特征组合起来,形成多尺度、层次化的特征;(3) 使用 softmax 回归分类器检索每个点属于预期语义类别的概率。所提出的方法针对两个公开可用的点云数据集进行了测试,以证明其性能并与其他现有方法产生的结果进行比较。实验结果在奥克兰数据集上实现了 96.89% 的总体准确率,在欧洲数据集上实现了 91.89% 的总体准确率,这是所考虑的方法中最高的。(3) 使用 softmax 回归分类器检索每个点属于预期语义类别的概率。所提出的方法针对两个公开可用的点云数据集进行了测试,以证明其性能并与其他现有方法产生的结果进行比较。实验结果在奥克兰数据集上实现了 96.89% 的总体准确率,在欧洲数据集上实现了 91.89% 的总体准确率,这是所考虑的方法中最高的。(3) 使用 softmax 回归分类器检索每个点属于预期语义类别的概率。所提出的方法针对两个公开可用的点云数据集进行了测试,以证明其性能并与其他现有方法产生的结果进行比较。实验结果在奥克兰数据集上实现了 96.89% 的总体准确率,在欧洲数据集上实现了 91.89% 的总体准确率,这是所考虑的方法中最高的。
更新日期:2018-12-10
down
wechat
bug