Coastal Wetland Vegetation Classification Using Pixel-Based, Object-Based and Deep Learning Methods Based on RGB-UAV
Jun-Yi Zheng,
Ying-Ying Hao,
Yuan-Chen Wang,
Si-Qi Zhou,
Wan-Ben Wu,
Qi Yuan,
Yu Gao,
Hai-Qiang Guo,
Xing-Xing Cai,
Bin Zhao
Affiliations
Jun-Yi Zheng
Ministry of Education Key Laboratory for Biodiversity Science and Ecological Engineering, National Observations and Research Station for Wetland Ecosystems of the Yangtze Estuary, and Shanghai Institute of EcoChongming (SIEC), Fudan University, Shanghai 200433, China
Ying-Ying Hao
Ministry of Education Key Laboratory for Biodiversity Science and Ecological Engineering, National Observations and Research Station for Wetland Ecosystems of the Yangtze Estuary, and Shanghai Institute of EcoChongming (SIEC), Fudan University, Shanghai 200433, China
Yuan-Chen Wang
Ministry of Education Key Laboratory for Biodiversity Science and Ecological Engineering, National Observations and Research Station for Wetland Ecosystems of the Yangtze Estuary, and Shanghai Institute of EcoChongming (SIEC), Fudan University, Shanghai 200433, China
Si-Qi Zhou
Ministry of Education Key Laboratory for Biodiversity Science and Ecological Engineering, National Observations and Research Station for Wetland Ecosystems of the Yangtze Estuary, and Shanghai Institute of EcoChongming (SIEC), Fudan University, Shanghai 200433, China
Wan-Ben Wu
Ministry of Education Key Laboratory for Biodiversity Science and Ecological Engineering, National Observations and Research Station for Wetland Ecosystems of the Yangtze Estuary, and Shanghai Institute of EcoChongming (SIEC), Fudan University, Shanghai 200433, China
Qi Yuan
Ministry of Education Key Laboratory for Biodiversity Science and Ecological Engineering, National Observations and Research Station for Wetland Ecosystems of the Yangtze Estuary, and Shanghai Institute of EcoChongming (SIEC), Fudan University, Shanghai 200433, China
Yu Gao
Key Laboratory of Fisheries Remote Sensing, East China Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Ministry of Agriculture and Rural Affairs, Shanghai 200090, China
Hai-Qiang Guo
Ministry of Education Key Laboratory for Biodiversity Science and Ecological Engineering, National Observations and Research Station for Wetland Ecosystems of the Yangtze Estuary, and Shanghai Institute of EcoChongming (SIEC), Fudan University, Shanghai 200433, China
Xing-Xing Cai
Ministry of Education Key Laboratory for Biodiversity Science and Ecological Engineering, National Observations and Research Station for Wetland Ecosystems of the Yangtze Estuary, and Shanghai Institute of EcoChongming (SIEC), Fudan University, Shanghai 200433, China
Bin Zhao
Ministry of Education Key Laboratory for Biodiversity Science and Ecological Engineering, National Observations and Research Station for Wetland Ecosystems of the Yangtze Estuary, and Shanghai Institute of EcoChongming (SIEC), Fudan University, Shanghai 200433, China
The advancement of deep learning (DL) technology and Unmanned Aerial Vehicles (UAV) remote sensing has made it feasible to monitor coastal wetlands efficiently and precisely. However, studies have rarely compared the performance of DL with traditional machine learning (Pixel-Based (PB) and Object-Based Image Analysis (OBIA) methods) in UAV-based coastal wetland monitoring. We constructed a dataset based on RGB-based UAV data and compared the performance of PB, OBIA, and DL methods in the classification of vegetation communities in coastal wetlands. In addition, to our knowledge, the OBIA method was used for the UAV data for the first time in this paper based on Google Earth Engine (GEE), and the ability of GEE to process UAV data was confirmed. The results showed that in comparison with the PB and OBIA methods, the DL method achieved the most promising classification results, which was capable of reflecting the realistic distribution of the vegetation. Furthermore, the paradigm shifts from PB and OBIA to the DL method in terms of feature engineering, training methods, and reference data explained the considerable results achieved by the DL method. The results suggested that a combination of UAV, DL, and cloud computing platforms can facilitate long-term, accurate monitoring of coastal wetland vegetation at the local scale.
