Airborne Laser Scanning Point Cloud Classification Using the DGCNN Deep Learning Method
| Autor: | Marda K. Fajari, E. Widyaningrum, Roderik Lindenbergh, Qian Bai |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
LiDAR
Computer science 0211 other engineering and technologies Point cloud 02 engineering and technology Convolutional neural network airborne point cloud deep learning classification accuracy assessment 0202 electrical engineering electronic engineering information engineering Feature (machine learning) lcsh:Science 021101 geological & geomatics engineering business.industry Deep learning Orthophoto Airborne point cloud Pattern recognition Classification Class (biology) Lidar Accuracy assessment General Earth and Planetary Sciences Graph (abstract data type) 020201 artificial intelligence & image processing lcsh:Q Artificial intelligence business |
| Zdroj: | Remote Sensing, 13(5) Remote Sensing, Vol 13, Iss 859, p 859 (2021) Remote Sensing; Volume 13; Issue 5; Pages: 859 |
| ISSN: | 2072-4292 |
| Popis: | Classification of aerial point clouds with high accuracy is significant for many geographical applications, but not trivial as the data are massive and unstructured. In recent years, deep learning for 3D point cloud classification has been actively developed and applied, but notably for indoor scenes. In this study, we implement the point-wise deep learning method Dynamic Graph Convolutional Neural Network (DGCNN) and extend its classification application from indoor scenes to airborne point clouds. This study proposes an approach to provide cheap training samples for point-wise deep learning using an existing 2D base map. Furthermore, essential features and spatial contexts to effectively classify airborne point clouds colored by an orthophoto are also investigated, in particularly to deal with class imbalance and relief displacement in urban areas. Two airborne point cloud datasets of different areas are used: Area-1 (city of Surabaya—Indonesia) and Area-2 (cities of Utrecht and Delft—the Netherlands). Area-1 is used to investigate different input feature combinations and loss functions. The point-wise classification for four classes achieves a remarkable result with 91.8% overall accuracy when using the full combination of spectral color and LiDAR features. For Area-2, different block size settings (30, 50, and 70 m) are investigated. It is found that using an appropriate block size of, in this case, 50 m helps to improve the classification until 93% overall accuracy but does not necessarily ensure better classification results for each class. Based on the experiments on both areas, we conclude that using DGCNN with proper settings is able to provide results close to production. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|