IMHE OpenIR
植被盖度遥感反演模型在高寒草原区的对比研究
Alternative TitleComparative study of remote sensing inversion model of Fractional Vegetation Cover in Alpine Grassland area
夏颖
Subtype硕士
Thesis Advisor范建容
2017
Degree Grantor中国科学院大学
Place of Conferral北京
Degree Discipline地图学与地理信息系统
Keyword植被盖度 高寒草原 光谱曲线 复合植被指数 像元二分模型 回归模型 改进的光谱梯度差模型
Abstract植被盖度是指示生态环境变化的基本指标,也是刻画地表植被覆盖的重要参数。在气候、水文、生态等领域有着广泛的应用。因此,快速、准确地获取植被盖度信息是非常有必要的。西藏高寒草原长期处于低温的环境条件中,导致其生长季节较短,草群比较稀疏、低矮,植被盖度较低,利用传统的遥感测量方法反演其植被盖度具有一定的难度。针对这一关键问题,研究选择了西南中部经南木林至申扎县的一条样带作为研究区,研究对象为低盖度的高寒草原。研究旨在提高高寒草原植被盖度的遥感反演精度,主要从以下三个方面进行了研究:一是如何充分利用野外采集的高光谱曲线,以构建适合于高寒草原的植被指数;二是针对研究区为干旱半干旱多山地区,以及高寒草原植被稀疏的特点,如何构建适合定量反演低植被盖度的复合植被指数;三是对比分析多种植被盖度遥感反演模型对高寒草原植被盖度的反演精度,以得到较适合反演稀疏高寒草原植被盖度的遥感反演模型。本研究所取得的主要成果如下:(1)利用野外采集的裸土光谱曲线,构建的转换型土壤调整植被指数(TSAVI),对稀疏高寒草原植被信息较为敏感。 (2)针对研究区为多山地区,且研究对象高寒草原分布稀疏的特点,引入了FCD模型中的裸土和阴影因子,结合已有的简单植被指数构建相应的复合植被指数。分别将简单植被指数与复合植被指数用于像元二分模型和回归模型中,从而反演高寒草原植被盖度。结果表明基于复合植被指数的模型对高寒草原植被盖度的反演精度高于简单植被指数。 (3)通过分析影像中低覆盖高寒草原以及高覆盖高寒草甸的光谱曲线特征,对光谱梯度差模型进行改进。然后利用面向对象规则的分类方法,提取出高寒草原植被信息。分别利用改进的光谱梯度差模型以及基于简单植被指数和复合植被指数的像元二分模型、回归模型对研究区高寒草原植被盖度进行反演,并以野外网格法实测的植被盖度作为基础数据,对反演的植被盖度进行精度的验证,以得到较适合于高寒草原植被盖度反演的遥感模型,并对高寒草原植被盖度进行分段统计。根据反演结果发现,像元二分模型对高寒草原植被盖度反演精度较其他两种模型更高,其中以基于TSAVI复合植被指数的像元二分模型最高,精度高达85.94%。
Other AbstractFractional Vegetation Cover (FVC) is the basic index of ecological environment change, and it is also an important parameter to describe the surface vegetation cover. It is widely used in the fields of climate, hydrology, ecology and so on. Therefore, it is very necessary to obtain FVC information quickly and accurately. But the Tibet Alpine Grassland in long-term low temperature conditions, resulting in a shorter growing season, grass group is relatively sparse, FVC is low, using the traditional measuring method by remote sensing to inversion the FVC of Alpine Grassland has a certain degree of difficulty. To solve this problem, a transect of alpine steppe from Shigatse City through Nanmulin County to Shenzha County in Tibet was selected as the research object,this study aims to improve the inversion precision of alpine grassland FVC, and mainly from the following three aspects: First is the research on how to make full use of the high spectral curve of field collection, to construct suitable vegetation index for alpine grassland; the sencond is for the study area is mountainous, arid and semi arid regions well, the alpine grassland vegetation sparse, for the characteristic, how to construct suitable complex vegetation index for quantitative inversion of FVC; The third is the comparative analysis the inversion accuracy of the alpine grassland FVC which is measured by various FVC inversion model, in order to get the most suitable remote sensing inversion model for alpine grassland FVC. The main results of this research are as follows: (1) Transformed Soil Adjusted Vegetation Index (TSAVI) was constructed by Measured Bare soil spectra data. And it’s sensitive to the low FVC information in semi arid area. (2) Because study area is a mountainous region, and the study object is sparse alpine grassland. For this characteristic, introducing the Bare soil index(BI) and shadow index(SI) factor of FCD model, BI and SI were combined with the existing single vegetation index to construct the corresponding composite vegetation index. The single vegetation index and composite vegetation index are used to construct the dimidiate pixel model and regression model, then use the dimidiate pixel model and regression model to inversion FVCof alpine grassland. The results show that the inversion precision of alpine grassland FVC which is measured by the model based on the composite vegetation index is higher than single. (3) The spectral gradient difference model was improved by analyzing the spectral characteristics of the Alpine Grassland and alpine meadow. Then, the alpine grassland vegetation information was extracted by the method of object-oriented classification. We used improved spectral gradient difference model , the dimidiate pixel model and regression model based on single vegetation index and composite vegetation index to inversion the FVC of the alpine grassland in the study area. In order to get the most suitable remote sensing inversion model for the alpine grassland FVC, the FVC which measured by grid method was used to verify the FVC inversion accuracy.Then we counted the alpine grassland FVC sectionally. According to the results, the inversion accuracy of the dimidiate pixel model is higher than the other two. And the dimidiate pixel model based on composite vegetation index VBSI(TSAVI) has the highest accuracy, it’s 85.94%.
Pages69
Language中文
Document Type学位论文
Identifierhttp://ir.imde.ac.cn/handle/131551/24566
Collection中国科学院水利部成都山地灾害与环境研究所
Affiliation中国科学院成都山地灾害与环境研究所
First Author Affilication中国科学院水利部成都山地灾害与环境研究所
Recommended Citation
GB/T 7714
夏颖. 植被盖度遥感反演模型在高寒草原区的对比研究[D]. 北京. 中国科学院大学,2017.
Files in This Item:
File Name/Size DocType Version Access License
植被盖度遥感反演模型在高寒草原区的对比研(2809KB)学位论文 开放获取CC BY-NC-SAView Application Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[夏颖]'s Articles
Baidu academic
Similar articles in Baidu academic
[夏颖]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[夏颖]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 植被盖度遥感反演模型在高寒草原区的对比研究.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.