基于多光谱影像和DEM的泥石流堆积扇识别研究——以白龙江流域武都段为例

发布时间：2018-10-25 09:13

【摘要】：选取中国泥石流多发区白龙江流域武都段作为研究区,在对该区域泥石流堆积扇的形态特征和堆积范围进行实地调绘的基础上,利用高分辨率影像(SPOT)进行目视解译,获得研究区部分泥石流堆积扇和非泥石流堆积区的分布范围,将其作为已知样本区。利用该区域多光谱遥感影像(ASTER)和数字高程模型(DEM),提取包含波段比和主分量的几十种特征指标。通过运用方差分析和聚类分析等方法对各指标进行分析计算,选取对区别泥石流堆积扇最具显著意义的指标进行输入,进而采用基于像元的分类方法识别泥石流堆积扇。得到如下结论:SPOT与ASTER融合影像的波段比、主分量指标可以有效地突出土壤岩石中的矿物成分,对泥石流堆积扇的识别具有显著意义;利用筛选出的遥感指标和地形指标作为输入,进行监督分类识别泥石流堆积扇,能够有效地将遥感指标和地形指标相结合,提取的堆积扇覆盖范围与实际情况较为接近。
[Abstract]:In this paper, the Wudu section of Bailong River basin is selected as the study area. On the basis of mapping the shape and range of debris flow accumulation fan in this area, high resolution image (SPOT) is used for visual interpretation. The distribution range of debris flow accumulation fan and non debris flow accumulation area in the study area is obtained and used as the known sample area. The multispectral remote sensing image (ASTER) and the digital elevation model (DEM),) are used to extract dozens of characteristic indexes including band ratio and principal component. By means of variance analysis and cluster analysis, the indexes of debris flow accumulation fan are analyzed and calculated, and the most significant indexes for distinguishing debris flow accumulation fan are selected for input, and then the pixel based classification method is used to identify debris flow accumulation fan. The conclusions are as follows: the band ratio of SPOT and ASTER fusion image, the principal component index can effectively protrude the mineral composition in soil and rock, and it is significant for the recognition of debris flow accumulation fan; The remote sensing index and topographic index can be used as input to monitor and classify debris flow accumulation fan, which can effectively combine remote sensing index with terrain index, and the area of coverage extracted from debris flow accumulation fan is close to the actual situation.
【作者单位】： 61243部队;
【分类号】：P237;P208;P642.23

【相似文献】