| 利用ETM+影像自动提取河流冲积扇的方法 |
| |
| 引用本文： | 杨树文,刘涛,冯光胜,谢飞.利用ETM+影像自动提取河流冲积扇的方法[J].测绘科学,2013,38(3):166-168. |
| |
| 作者姓名： | 杨树文 刘涛 冯光胜 谢飞 |
| |
| 作者单位： | 1. 兰州交通大学测绘与地理信息学院,兰州,730070
2. 铁道第四勘察设计院,武汉,430063 |
| |
| 基金项目： | 中铁第四勘查设计院集团有限公司基金项目 |
| |
| 摘 要： | 本文在分析冲积扇与其他地物光谱特征差异的基础上,针对Landsat ETM+影像中红光波段与近红外波段的比值能够增大冲积扇与其他地物间差异的特征及地形阴影在蓝绿光波段亮度值降低速率差异较大的特征,基于比值运算、差值运算,构建了冲积扇指数;并利用该模型阈值自动选取算法将冲积扇信息从其他地物及阴影中分离出来,然后根据数学形态学膨胀滤波算法进行空洞填充。经过实验表明,该方法能够高效自动地提取华南山区的冲积扇信息。
|
| 关 键 词： | 冲积扇 光谱特征 自动提取 阈值 滤波 |
Automatic extraction method of alluvial fan based on ETM + images |
| |
| YANG Shu-wen,LIU Tao,FENG Guang-sheng,XIE Fei.Automatic extraction method of alluvial fan based on ETM + images[J].Science of Surveying and Mapping,2013,38(3):166-168. |
| |
| Authors: | YANG Shu-wen LIU Tao FENG Guang-sheng XIE Fei |
| |
| Institution: | (①Faculty of Geomatics,Physics&Software Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China;②Fourth Survey and Design Institute of China Railway,Wuhan 430063,China) |
| |
| Abstract: | An automatic extraction approach for alluvial fan based on ETM+ images was put forward in the paper.Firstly the spectral features differences between alluvial fan and other land features were analyzed;in view of the band ratio between red band and near-infrared band could increase the differences between alluvial fan and other surface features,and the decrease rate of bright values of shadow in blue&green bands are significantly different,the Alluvial Fan Index(AFI) was build based on ratio operation and difference operation.Secondly,using this model,combined with algorithm of automatic threshold extraction,alluvial fan was separated from other surface features and shadows,then dilation filtering algorithm of mathematical morphology was allied.The analysis and comparison through the experiment showed that the proposed approach could effectively extract the alluvial fan of the South China mountain area with high precision. |
| |
| Keywords: | alluvial fan spectral features automatic extraction threshold filtering |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
