Mapping Mangrove Density from Rapideye Data in Central America

Nguyen Thanh Son, Chi Farn Chen, Cheng Ru Chen

研究成果: 雜誌貢獻期刊論文同行評審

10 引文 斯高帕斯(Scopus)

摘要

Mangrove forests provide a wide range of socioeconomic and ecological services for coastal communities. Extensive aquaculture development of mangrove waters in many developing countries has constantly ignored services of mangrove ecosystems, leading to unintended environmental consequences. Monitoring the current status and distribution of mangrove forests is deemed important for evaluating forest management strategies. This study aims to delineate the density distribution of mangrove forests in the Gulf of Fonseca, Central America with Rapideye data using the support vector machines (SVM). The data collected in 2012 for density classification of mangrove forests were processed based on four different band combination schemes: scheme-1 (bands 1-3, 5 excluding the red-edge band 4), scheme-2 (bands 1-5), scheme-3 (bands 1-3, 5 incorporating with the normalized difference vegetation index, NDVI), and scheme-4 (bands 1-3, 5 incorporating with the normalized difference red-edge index, NDRI). We also hypothesized if the obvious contribution of Rapideye red-edge band could improve the classification results. Three main steps of data processing were employed: (1) data pre-processing, (2) image classification, and (3) accuracy assessment to evaluate the contribution of red-edge band in terms of the accuracy of classification results across these four schemes. The classification maps compared with the ground reference data indicated the slightly higher accuracy level observed for schemes 2 and 4. The overall accuracies and Kappa coefficients were 97% and 0.95 for scheme-2 and 96.9% and 0.95 for scheme-4, respectively.

原文???core.languages.en_GB???
頁(從 - 到)211-220
頁數10
期刊Open Geosciences
9
發行號1
DOIs
出版狀態已出版 - 2017

指紋

深入研究「Mapping Mangrove Density from Rapideye Data in Central America」主題。共同形成了獨特的指紋。

引用此