English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Point cloud classification plays an important role in a wide range ofairborne light detection and ranging (LiDAR) applications, such as topographicmapping, forest monitoring, power line detection, and road detection. However,due to the sensor noise, high redundancy, incompleteness, and complexity ofairborne LiDAR systems, point cloud classification is challenging. In thispaper, we proposed a directionally constrained fully convolutional neuralnetwork (D-FCN) that can take the original 3D coordinates and LiDAR intensityas input; thus, it can directly apply to unstructured 3D point clouds forsemantic labeling. Specifically, we first introduce a novel directionallyconstrained point convolution (D-Conv) module to extract locally representativefeatures of 3D point sets from the projected 2D receptive fields. To make fulluse of the orientation information of neighborhood points, the proposed D-Convmodule performs convolution in an orientation-aware manner by using adirectionally constrained nearest neighborhood search. Then, we designed amultiscale fully convolutional neural network with downsampling and upsamplingblocks to enable multiscale point feature learning. The proposed D-FCN modelcan therefore process input point cloud with arbitrary sizes and directlypredict the semantic labels for all the input points in an end-to-end manner.Without involving additional geometry features as input, the proposed methodhas demonstrated superior performance on the International Society forPhotogrammetry and Remote Sensing (ISPRS) 3D labeling benchmark dataset. Theresults show that our model has achieved a new state-of-the-art level ofperformance with an average F1 score of 70.7%, and it has improved theperformance by a large margin on categories with a small number of points (suchas powerline, car, and facade).

Directionally Constrained Fully Convolutional Neural Network For Airborne Lidar Point Cloud Classification