| 面向对象的高分辨率影像城市建筑物提取 |
| |
| 引用本文： | 乔程,骆剑承,吴泉源,沈占锋,王宏.面向对象的高分辨率影像城市建筑物提取[J].地理与地理信息科学,2008,24(5). |
| |
| 作者姓名： | 乔程 骆剑承 吴泉源 沈占锋 王宏 |
| |
| 作者单位： | 1. 中国科学院遥感应用研究所,北京,100101
2. 山东师范大学人口·资源与环境学院,山东,济南,250014 |
| |
| 基金项目： | 国家高技术研究发展计划(863计划) |
| |
| 摘 要： | 依据高分辨率遥感影像的特点,采用面向对象方法提出"自下而上"的多尺度分割方法,即按照由小尺度分割至大尺度分割的顺序,并结合掩膜操作对山东师范大学IKONOS影像进行了建筑物提取试验。结果表明,该方法较传统的"自上而下"的尺度分割方法精度有显著改善,其分类结果形状较为规整,更接近实际地物。
|
| 关 键 词： | 高分辨率遥感 面向对象 城市建筑物 多尺度分割 |
Object-Oriented Method Based Urban Building Extraction from High Resolution Remote Sensing Image |
| |
| QIAO Cheng,LUO Jian-cheng,WU Quan-yuan,SHEN Zhan-feng,WANG Hong.Object-Oriented Method Based Urban Building Extraction from High Resolution Remote Sensing Image[J].Geography and Geo-Information Science,2008,24(5). |
| |
| Authors: | QIAO Cheng LUO Jian-cheng WU Quan-yuan SHEN Zhan-feng WANG Hong |
| |
| Abstract: | Development of high resolution remote sensing makes it become a key factor in areas of surveying and mapping,urban planning,resources and environment,etc.In the same time,it is badly needed in change detection of urban buildings imposed by city's accelerated development.High-resolution remote sensing images have many more spatial and textural characteristics than low-resolution data except spectral characteristics.Thus,using multi-scale segmentation based on object-oriented method,combined with mask operation to extract buildings from IKONOS image of Shandong normal university,that is,do small-scale segmentation first,then large-scale.The result shows that this method is more efficient,and it provides a new technical approach for the information extraction from high resolution images. |
| |
| Keywords: | high resolution remote sensing object-oriented urban building multi-scale segmentation |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
