文献类型：期刊文献

英文题名：Forest Canopy Closure Estimation in Greater Khingan Forest Based on Gf-2 Data

作者：Sun, Shanshan[1] Li, Zengyuan[1] Tian, Xin[1] Gao, Zhihai[1] Wang, Chongyang[1] Gu, Chengyan[2]

第一作者：Sun, Shanshan

通信作者：Li, Zengyuan

机构：[1] Chinese Academy of Forestry, Institute of Forest Resource Information Techniques, Beijing, 100091, China; [2] Forestry Planning and Design Academy of Forest Products Industry, Beijing, 100010, China

年份：2019

起止页码：6640-6643

外文期刊名：International Geoscience and Remote Sensing Symposium (IGARSS)

收录：EI(收录号：20200208026074)

语种：英文

外文关键词：Ecosystems - Optical radar - Mean square error - Remote sensing - Textures - Forestry

摘要：Forest canopy closure (FCC) is an important factor to assess the quality of forest resources, and to understand the characteristics of forest change, which supports forest ecosystem management. The Chinese high-resolution satellite-2 (Gaofen-2, GF-2) image covering the Genhe Forest Reserve located at the Great Khingan of Inner Mongolia was firstly segmented by object-oriented technology and then the local FCC was estimated by the support vector machine (SVM) based on the GF-2's spectral, normalized vegetation index (NDVI), texture and other auxiliary information. The FCC estimates from the airborne LiDAR point cloud data with high density were used for the cross-validation. The result showed that the coefficient of determination (R2) between LiDAR and GF2 results was up to 0.65 and the root mean square error (RMSE) is 0.12. It indicated that it is feasible to estimate FCC by using the GF-2 images based on the object-oriented classification method. ? 2019 IEEE.