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：2000
卷号：XXXIII Part B7(/1-4)
页码：1588-1592
出版社：Copernicus Publications
摘要：The need for reliable and up-to-date Pan European Land Use (LU) and Land Cover (LC) information is well known. Some steps have already been taken; one of them in the Pan-European Land Cover Monitoring project (PELCOM) in the EC 4th Framework Program. A concise classification methodology was established for mapping major land cover types (e.g. forest, grassland, and arable land) using a regional and integrated approach of NOAA-AVHRR satellite data and ancillary information. It is the sensible integration of the numerously available sources of LU/LC related information that is expected to greatly enhance the value of the information derived from satellites. The available ancillary LU/LC information ranges from continuos topographic information (altitude, distance from the sea, latitude), to climatic maps, to existing thematic land cover maps (SEI map, CORINE, wetlands map), all providing complementary information on the actual and potential land cover. Combining the ancillary sources of information with the processed land cover signatures derived from, for instance the NOAA/AVHRR satellites, it becomes feasible to produce more reliable land cover information. In the PELCOM project, a technique has been developed and tested for integrating various sources of LU/LC information. The most critical part of this technique is to capture the qualitative knowledge that is typically available; valuable knowledge of both the reliability of the information and of the content of the available sources. What does the presence of a particular source class infer about the likelihood of the target classes. The challenge is to convert this
关键词：Landcover; Remote Sensing; Fusion; Integration; Land Use
