文章基本信息

标题：CONGO BASIN FOREST COVER CHANGE ESTIMATE FOR 1990, 2000 AND 2005 BY LANDSAT INTERPRETATION USING AN AUTOMATED OBJECT-BASED PROCESSING CHAIN
本地全文：下载
作者：C. Ernsta ; A. Verhegghen ; C. Bodart 等
期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2010
卷号：XXXVIII - 4/C7
出版社：Copernicus Publications
摘要：Central Africa contains the second largest area of contiguous moist tropical forest of the world. In the framework of the Observatory of Forests in Central Africa (OFAC) and the Forest Resources Assessment (FRA-2010) led by Food and Agricultural Organization of the United Nations (FAO), each country is invited to provide an estimation of forest cover change for years 1990-2000-2005 (and later 2010). In this context, developing efficient methods to detect forest changes by processing remote sensing data is more than ever a challenging need. At the moment, only satellite images can provide enough information on processes such as deforestation at the scale of Congo basin. An automatic method has been designed to map and quantify forest change in Central Africa. 1168 subsets of 20 x 20 km of 30 m resolution Landsat or Aster are required to cover the different countries. The different steps of the method are a (1) multi-date segmentation, applied on each extracts triplet (1990-2000 and 2005), (2) an unsupervised classification and an (3) automatic pre-labeling. (4) Change is detected by a statistical object-based method. (5) The involvement of national experts is an essential part of the intepretation process. The OFAC team together with Joint Research Centre (JRC-EU) and FAO invited in Kinshasa (September 2009) 13 national experts from the 6 countries of the Congo Basin to validate the land cover mapping and change maps. National experts check, and change if needed, the pre-interpretation of each sample using an object-based validation tool developed by the JRC. This unique exercise estimates not only deforestation and reforestation but also degradation and regeneration which are particularly important in Central Africa. These results are expected to contribute to the discussion on the reduction of CO2 emissions from deforestation and forest degradation (UN-REDD)
关键词：Congo Basin; forest cover; REDD; land cover; land cover change; object-based validation; high spatial resolution