【作者】 时王侠；

【导师】 余明；

【作者基本信息】 福建师范大学 ， 地图学与地理信息系统， 2008， 硕士

【摘要】 GIS的应用提高了遥感影像的数据提取和分析能力,遥感影像信息的提取和分类是GIS在土地资源应用中最主要的信息源。解决遥感影像识别问题并满足一定的精度,是遥感影像分析中的一个关键问题,具有十分重要的意义。本文选取地物类型较复杂的福州市的城乡结合区的一小块区域为试验区,并集成遥感影像的光谱特征、纹理特征及地形特征构建了多源空间数据库,利用基于C4.5算法、基于粗糙集理论及基于粗糙集理论和C4.5算法相结合的方法,从空间数据库的训练样本数据集中挖掘分类规则,进行分类实验,并分别与传统的最大似然分类法进行比较与分析。研究结果表明,基于知识分类方法的分类精度都明显高于传统的最大似然法,其中基于粗糙集和C4.5算法相结合的分类方法的分类效果最好,并不同程度的弥补了单一的基于粗糙集理论和基于C4.5算法分类中的不足。因而基于多种算法相结合的分类方法,可以充分发挥其优点,并弥补分类中各自的不足,是促进基于知识的遥感影像分类方法在土地利用分类中广泛应用的一项有效手段。更多还原

【Abstract】 The Application of GIS enhances the data extraction and analysis ability of remote sensing images, and the information extraction and classification of remote sensing image is the most important source of information in the land resources application of GIS. However, how to distinguish kinds of Images and meet the accuracy to a certain extent is a key problem in remote sensing image research, which takes on very important significance.The paper takes a small plot in the outskirts of Fuzhou with complicated land use in Fujian Province as study area. The classification rules are discovered from the training samples of the spatial database based on C4.5 algorithm, Rough Sets and both with together, which integrate spectral characteristics, texture characteristics and terrain features of remote sensing image to build a multi-source spatial database. And the classification tests are carried out based on the rules. The traditional maximum likelihood classification is also compared to check the classification accuracies. The results have suggested that the accuracy of classification based on knowledge is markedly higher than the traditional maximum likelihood classification. The combination of Rough Sets and C4.5 algorithm is the best, which can make up for the shorts based on Rough Sets or C4.5 algorithm to a different extent. So the classification method based on multi-algorithm, which can take a good advantage of kinds of classification algorithm and make up for the shorts respectively, is a good way to promote the wide application of the classification methods of remote sensing image based on knowledge in land use classification.更多还原