Abstract:LiDAR point clouds are rich in spatial information and can effectively express the size, shape, position, and direction of objects; thus, they have the advantage of high spatial utilization. The point cloud focuses on describing the shape of the external surface of the object itself and will not store useless redundant information to describe the occupation. Therefore, point clouds have become the research focus of 3D data models and are widely used in large-scale scene reconstruction, virtual reality, digital elevation model production, and other fields. Since point clouds have various characteristics, such as disorder, density inconsistency, unstructuredness, and incomplete information, point cloud classification is still complex and challenging. To realize the semantic classification of LiDAR point clouds in complex scenarios, this paper proposes the integration of normal vector features into an atrous convolution residual network. Based on the RandLA-Net network structure, the proposed network integrates the atrous convolution into the residual module to extract global and local features of the point clouds. The atrous convolution can learn more valuable point cloud feature information by expanding the receptive field. Then, the point cloud normal vector is embedded in the local feature aggregation module of the RandLA-Net network to extract local semantic aggregation features. The improved local feature aggregation module can merge the deep features of the point cloud and mine the fine-grained information of the point cloud to improve the model’s segmentation ability in complex scenes. Finally, to resolve the imbalance of the distribution of the various categories of point clouds, the original loss function is optimized by adopting a reweighted method to prevent overfitting so that the network can focus on small target categories in the training process to effectively improve the classification performance. Through the experimental analysis of a Vaihingen (Germany) urban 3D semantic dataset from the ISPRS website, it is verified that the proposed algorithm has a strong generalization ability. The overall accuracy (OA) of the proposed algorithm on the Vaihingen urban 3D semantic dataset reached 97.9%, and the average reached 96.1%. Experiments show that the proposed algorithm fully exploits the semantic features of point clouds and effectively improves the accuracy of point cloud classification.

Airborne LiDAR point cloud classification using PointNet++ network with full neighborhood features

PointNet++ Network Architecture with Individual Point Level and Global Features on Centroid for ALS Point Cloud Classification

Point Cloud Completion Cascade Optimization Network Based on Feature Fusion

Multispectral LiDAR Point Cloud Classification Using SE-PointNet++

MLF-PointNet++: A Multifeature-Assisted and Multilayer Fused Neural Network for LiDAR-UAS Point Cloud Classification in Estuarine Areas

Classifying airborne LiDAR point clouds via deep features learned by a multi-scale convolutional neural network

Airborne LiDAR Point Cloud Classification with Graph Attention Convolution Neural Network

CBF-Net: An Adaptive Context Balancing and Feature Filtering Network for Point Cloud Classification

Density-Aware Convolutional Networks with Context Encoding for Airborne LiDAR Point Cloud Classification

Tree Species Classification Based on PointNet++ and Airborne Laser Survey Point Cloud Data Enhancement

Tree Species Classification of Backpack Laser Scanning Data Using the PointNet++ Point Cloud Deep Learning Method

GSV-NET: A Multi-Modal Deep Learning Network for 3D Point Cloud Classification

ALS Point Cloud Classification by Integrating an Improved Fully Convolutional Network into Transfer Learning with Multi-Scale and Multi-View Deep Features

Integrating Normal Vector Features into an Atrous Convolution Residual Network for LiDAR Point Cloud Classification

BiPointNet: Binary Neural Network for Point Clouds

Point Set Multi-Level Aggregation Feature Extraction Based on Multi-Scale Max Pooling and LDA for Point Cloud Classification

Semantic segmentation of sparse irregular point clouds for leaf/wood discrimination

Airborne Laser Scanning Point Cloud Classification Using the DGCNN Deep Learning Method

A Comparison of Deep Learning Methods for Airborne Lidar Point Clouds Classification