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：2006
卷号：XXXVI-4/C42
出版社：Copernicus Publications
摘要：Remote sensing devices have collected a considerable amount of data about the nature of the Earth's surface. How to extract high precise information from remotely sensed data, however, is one of the key issues on research and application of remote sensing. Remote sensing image classification and analysis have become the context in which special information extraction, dynamic change forecast, thematic mapping and building remote sensing dataset will be implemented. Based on the principle of remote sensing image classification, the paper roundly discusses all kinds of conservative methods of supervised classification and unsupervised classification of remote sensing image, and simply appraises their advantage and disadvantage, and suitable occasions. The newest methods of computerized remote sensing image classification are discussed with the attempt to bring out the trend of remote sensing image classification. In the end, this paper proposes a new classification method based on the multi-resolution hierarchy in order to have new breakthrough for the research of the classification method of remote sensing image
