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：496-503
出版社：Copernicus Publications
摘要：Humid areas and natural wetlands are among the most important ecological resources. Management of these areas is becoming a mandatory element in nature conservation. Until today, the collection of relevant information on humid areas is mainly based on field work. Use of digital imagery has not yet evolved beyond an experimental stage. In order to set up a processing chain for remote sensing products that help in wetland management, the MANHUMA project has been funded and launched by the European Union. Based on a report on customer needs and according to three given management functions, three remote sensing products have been defined. In order to evaluate the potential of future satellite missions, it was our task to process data of the DAEDALUS airborne sensor. In this paper a land cover map, based on a class list defined by the customers, will be presented as the base product for all three management functions. Results show that the main limit is not only to be found on the technical side of the sensor, but also in nature's great variety. Soil parameters like moisture or fertility often vary within distances of 10–20 m, affecting the phenological status of vegetation significantly. Overall classification results show that an accurate delineation of the different land cover types was possible at the sensor's resolution. A good characterisation of all test sites through land cover classes that are relevant to nature conservation measures has been achieved
关键词：Remote Sensing; User requirements; High-resolution data/images; Multi-spectral data; Classification; ; Land use / Land cover; Performance Analysis; International co-operation
