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：2007
卷号：XXXVI-2/C43
出版社：Copernicus Publications
摘要：Countryside agencies in the UK are interested in how to generate a range of Boolean maps whilst controlling uncertainties in ways that are appropriate to different landscape questions. This work illustrates the different effects of Bayes and Dempster-Shafer in combining ancillary information to augment remote sensing analyses. The example of translating between different habitat classifi- cations is used (Phase I, Priority and Annex I) to test the hypothesis that it is possible to move beyond from the concept of crisp mappings to one based on bounded belief. The results highlight the importance of data scale and grain of process in the development context-sensitive Boolean maps and indicate that conservation managers need to be able to identify the 'best' decision in the face of uncertainty relating to the nature of the object or process and scale
