A Multiscale Multi-Feature Deep Learning Model for Airborne Point-Cloud Semantic Segmentation

In point-cloud scenes, semantic segmentation is the basis for achieving an understanding of a 3D scene. The disorderly and irregular nature of 3D point clouds makes it impossible for traditional convolutional neural networks to be applied directly, and most deep learning point-cloud models often suffer from an inadequate utilization of spatial information and of other related point-cloud features. Therefore, to facilitate the capture of spatial point neighborhood information and obtain better performance in point-cloud analysis for point-cloud semantic segmentation, a multiscale, multi-feature PointNet (MSMF-PointNet) deep learning point-cloud model is proposed in this paper. MSMF-PointNet is based on the classical point-cloud model PointNet, and two small feature-extraction networks called Mini-PointNets are added to operate in parallel with the modified PointNet; these additional networks extract multiscale, multi-neighborhood features for classification. In this paper, we use the spherical neighborhood method to obtain the local neighborhood features of the point cloud, and then we adjust the radius of the spherical neighborhood to obtain the multiscale point-cloud features. The obtained multiscale neighborhood feature point set is used as the input of the network. In this paper, a cross-sectional comparison analysis is conducted on the Vaihingen urban test dataset from the single-scale and single-feature perspectives.

[1]  Jin-Lin Cai,et al.  Point cloud classification network based on self-attention mechanism , 2022, Comput. Electr. Eng..

[2]  Qingming Huang,et al.  C2FNet: A Coarse-to-Fine Network for Multi-View 3D Point Cloud Generation , 2022, IEEE Transactions on Image Processing.

[3]  Yindi Zhao,et al.  Multi-Feature Aggregation for Semantic Segmentation of an Urban Scene Point Cloud , 2022, Remote. Sens..

[4]  S. Tang,et al.  SEMANTIC SEGMENTATION OF INDOOR 3D POINT CLOUDS BY JOINT OPTIMIZATION OF GEOMETRIC FEATURES AND NEURAL NETWORKS , 2022, ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences.

[5]  Wenbing Tao,et al.  SVASeg: Sparse Voxel-Based Attention for 3D LiDAR Point Cloud Semantic Segmentation , 2022, Remote. Sens..

[6]  Yongyang Xu,et al.  RG-GCN: A Random Graph Based on Graph Convolution Network for Point Cloud Semantic Segmentation , 2022, Remote. Sens..

[7]  Guo Zhang,et al.  AGFP-Net: Attentive geometric feature pyramid network for land cover classification using airborne multispectral LiDAR data , 2022, Int. J. Appl. Earth Obs. Geoinformation.

[8]  Yuntao He,et al.  Adaptive deep learning-based neighborhood search method for point cloud , 2022, Scientific Reports.

[9]  Shuo Shi,et al.  Target Classification of Similar Spatial Characteristics in Complex Urban Areas by Using Multispectral LiDAR , 2022, Remote. Sens..

[10]  Jie Yang,et al.  Graph Attention Feature Fusion Network for ALS Point Cloud Classification , 2021, Sensors.

[11]  Gang Yu,et al.  BiSeNet V2: Bilateral Network with Guided Aggregation for Real-Time Semantic Segmentation , 2020, International Journal of Computer Vision.

[12]  Francesco De Luca,et al.  A Computational Workflow for Generating A Voxel-Based Design Approach Based on Subtractive Shading Envelopes and Attribute Information of Point Cloud Data , 2020, Remote. Sens..

[13]  Luis Riazuelo,et al.  MiniNet: An Efficient Semantic Segmentation ConvNet for Real-Time Robotic Applications , 2020, IEEE Transactions on Robotics.

[14]  Luis Riazuelo,et al.  3D-MiniNet: Learning a 2D Representation From Point Clouds for Fast and Efficient 3D LIDAR Semantic Segmentation , 2020, IEEE Robotics and Automation Letters.

[15]  Alexandre Boulch.  ConvPoint: Continuous convolutions for point cloud processing , 2019, Comput. Graph..

[16]  侯向丹 Hou Xiangdan,et al.  3D Point Cloud Classification and Segmentation Model Based on Graph Convolutional Network , 2020, Laser & Optoelectronics Progress.

[17]  Mingwei Wang,et al.  A Building Extraction Approach Based on the Fusion of LiDAR Point Cloud and Elevation Map Texture Features , 2019, Remote. Sens..

[18]  Zhe Zhu,et al.  Understanding an urbanizing planet: Strategic directions for remote sensing , 2019, Remote Sensing of Environment.

[19]  Faming Shao,et al.  Energy-Efficient Gabor Kernels in Neural Networks with Genetic Algorithm Training Method , 2019, Electronics.

[20]  Wanshou Jiang,et al.  Segmentation and Multi-Scale Convolutional Neural Network-Based Classification of Airborne Laser Scanner Data , 2018, Sensors.

[21]  José García Rodríguez,et al.  A survey on deep learning techniques for image and video semantic segmentation , 2018, Appl. Soft Comput..

[22]  Ruibin Zhao,et al.  Classifying airborne LiDAR point clouds via deep features learned by a multi-scale convolutional neural network , 2018, Int. J. Geogr. Inf. Sci..

[23]  Dong Chen,et al.  LiDAR Point Clouds to 3-D Urban Models$:$ A Review , 2018, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[24]  Carl Salvaggio,et al.  A Fully Convolutional Network for Semantic Labeling of 3D Point Clouds , 2017, ISPRS Journal of Photogrammetry and Remote Sensing.

[25]  Xiaoye Liu,et al.  Airborne LiDAR for DEM generation: some critical issues , 2008 .