TÀI LIỆU SỐ MỚI CẬP NHẬT
HỖ TRỢ TRỰC TUYẾN
THỐNG KÊ TRUY CẬP
  • Đang trực tuyến: 153
  • Tổng lượt truy cập: 6404256
Monitoring mangrove biomass change in Vietnam using SPOT images and an object-based approach combined with machine learning algorithms.
12/07/19 02:16PM
Lien T. H. Pham and Lars Brabyn. ISPRS Journal of Photogrammetry and Remote Sensing, 2017, volume 128, pp. 86-97.
Abstract: Mangrove forests are well-known for their provision of ecosystem services and capacity to reduce carbon dioxide concentrations in the atmosphere. Mapping and quantifying mangrove biomass is useful for the effective management of these forests and maximizing their ecosystem service performance. The objectives of this research were to model, map, and analyse the biomass change between 2000 and 2011 of mangrove forests in the Cangio region in Vietnam. SPOT 4 and 5 images were used in conjunction with object-based image analysis and machine learning algorithms. The study area included natural and planted mangroves of diverse species. After image preparation, three different mangrove associations were identified using two levels of image segmentation followed by a Support Vector Machine classifier and a range of spectral, texture and GIS information for classification. The overall classification accuracy for the 2000 and 2011 images were 77.1% and 82.9%, respectively. Random Forest regression algorithms were then used for modelling and mapping biomass. The model that integrated spectral, vegetation association type, texture, and vegetation indices obtained the highest accuracy (R2adj = 0.73). Among the different variables, vegetation association type was the most important variable identified by the Random Forest model. Based on the biomass maps generated from the Random Forest, total biomass in the Cangio mangrove forest increased by 820,136 tons over this period, although this change varied between the three different mangrove associations.
TRANG CHỦ GIỚI THIỆU SẢN PHẨM - DỊCH VỤ PHÁT HÀNH SÁCH CSDL Thông tin TIN HOẠT ĐỘNG ĐIỂM BÁO LIÊN HỆ