| 以SPOT5遥感数据为例比较基于像素与面向对象的分类方法 |
| |
| 引用本文： | 胡荣明,魏曼,杨成斌,贺俊斌.以SPOT5遥感数据为例比较基于像素与面向对象的分类方法[J].遥感技术与应用,2012,27(3):366-371. |
| |
| 作者姓名： | 胡荣明 魏曼 杨成斌 贺俊斌 |
| |
| 作者单位： | (西安科技大学测绘学院,陕西 西安 710054) |
| |
| 摘 要： | 以SPOT5高空间分辨率遥感影像为研究数据,采用面向对象的遥感影像分类方法,将基于像素的分类精度与基于影像对象的分类精度进行对比分析。结果表明:基于像素的分类方法在处理高空间分辨率遥感影像时具有一定的局限性,而面向对象的分类方法充分利用了影像中的光谱信息、几何结构、空间信息及上下文信息,在高空间分辨率遥感影像分类方面更具有优势,有良好的应用前景。
|
| 关 键 词： | SPOT5 高分辨率影像 像素 面向对象 分类 |
| 收稿时间： | 2011-09-08 |
Taking SPOT5 Remote Sensing Data for Example to Compare Pixel-based and Object-oriented Classification |
| |
| Hu Rongming,Wei Man,Yang Chengbin,He Junbin.Taking SPOT5 Remote Sensing Data for Example to Compare Pixel-based and Object-oriented Classification[J].Remote Sensing Technology and Application,2012,27(3):366-371. |
| |
| Authors: | Hu Rongming Wei Man Yang Chengbin He Junbin |
| |
| Affiliation: | (College of Geomatics,Xian University of Science and Technology,Xian 710054,China) |
| |
| Abstract: | The high spatial resolution remote sensing image SPOT5 for study data,using object-oriented classification method,this paper compared the classification accuracy of pixel-based and object-based.The results show that pixel-based classification has certain limitation and object-oriented classification method takes advantage of the spectral information,geometry information,spatial information,even the context information in processing high resolution remote sensing image.Therefore,object-oriented classification method has more advantages and clear application prospect. |
| |
| Keywords: | SPOT5 High spatial resolution image Pixel Object-oriented Classification |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《遥感技术与应用》浏览原始摘要信息 |
|
点击此处可从《遥感技术与应用》下载全文 |
|
