Extracting Urban Impervious Surface from WorldView-2 and Airborne LiDAR Data Using 3D Convolutional Neural Networks
- PDF / 8,569,251 Bytes
- 12 Pages / 595.276 x 790.866 pts Page_size
- 42 Downloads / 214 Views
RESEARCH ARTICLE
Extracting Urban Impervious Surface from WorldView-2 and Airborne LiDAR Data Using 3D Convolutional Neural Networks Zhongchang Sun1,2 • Xiangwei Zhao3
•
Mengfan Wu4 • Cuizhen Wang5
Received: 19 July 2018 / Accepted: 15 November 2018 / Published online: 7 December 2018 Ó The Author(s) 2018
Abstract The urban impervious surface has been recognized as a key quantifiable indicator in assessing urbanization and its environmental impacts. Adopting deep learning technologies, this study proposes an approach of three-dimensional convolutional neural networks (3D CNNs) to extract impervious surfaces from the WorldView-2 and airborne LiDAR datasets. The influences of different 3D CNN parameters on impervious surface extraction are evaluated. In an effort to reduce the limitations from single sensor data, this study also explores the synergistic use of multi-source remote sensing datasets for delineating urban impervious surfaces. Results indicate that our proposed 3D CNN approach has a great potential and better performance on impervious surface extraction, with an overall accuracy higher than 93.00% and the overall kappa value above 0.89. Compared with the commonly applied pixel-based support vector machine classifier, our proposed 3D CNN approach takes advantage not only of the pixel-level spatial and spectral information, but also of texture and feature maps through multi-scale convolutional processes, which enhance the extraction of impervious surfaces. While image analysis is facing large challenges in a rapidly developing big data era, our proposed 3D CNNs will become an effective approach for improved urban impervious surface extraction. Keywords WorldView-2 Airborne light detection and ranging (LiDAR) Impervious surface Convolutional neural networks (CNNs) Support vector machine (SVM) JEL Classification 42020
Introduction
& Xiangwei Zhao [email protected]; [email protected] 1
Laboratory of Digital Earth Science, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100094, China
2
Key Laboratory Earth Observation Hainan Province, Sanya Institute of Remote Sensing, Sanya 572029, Hainan, China
3
Shandong Province ‘‘3S’’ Engineering Research Center, Shandong University of Science and Technology, Qingdao 266590, China
4
Faculty of Geographical Science, Beijing Normal University, Beijing 100875, China
5
Department of Geography, University of South Carolina, Columbia, SC 29208, USA
Impervious surfaces are usually defined as the entirety of impermeable surfaces such as roads, buildings, parking lots, and other urban infrastructures, where water cannot infiltrate through the ground (Sun et al. 2011). Urbanization results in the increase in impervious surfaces, which in turn casts great impacts on urban environmental problems such as increased urban heat islands (Ma et al. 2016), surface runoff (Sun et al. 2014), water contamination (Kim et al. 2016), and air pollution (Touchaei et al. 2016). Facing rapid urbanization all over the world, these environ
Data Loading...
