文献类型：期刊文献

中文题名：基于SPOT-5光谱和纹理信息的湘西森林生态功能指数遥感预测模型构建

英文题名：Estimation of the Forest Ecological Function Index in Western Hunan Using the Spectral and Textural Information Derived from SPOT-5 Satellite Images

作者：李晗[1] 陈新云[2] 白彦锋[3] 姜春前[3] 孟京辉[1]

第一作者：李晗

机构：[1]北京林业大学省部共建森林培育与保护教育部重点实验室,北京100083;[2]国家林业局调查规划设计院,北京100714;[3]中国林业科学研究院林业研究所,北京100091

年份：2019

卷号：34

期号：5

起止页码：147-153

中文期刊名：西北林学院学报

外文期刊名：Journal of Northwest Forestry University

收录：CSTPCD;;北大核心:【北大核心2017】；CSCD:【CSCD_E2019_2020】；

基金：国家重点研发计划(2017YFC0505604)

语种：中文

中文关键词：森林生态功能指数;SPOT-5;Pearson相关分析;全子集回归

外文关键词：forest ecological function index;SPOT-5;Pearson s correlation analysis;all subsets multiple linear regression

分类号：S771.8

摘要：以湘西区域SPOT-5遥感影像为基础,提取样地纹理和光谱信息,以一类调查数据的生态指数为因变量,所对应的纹理和光谱指数为自变量,采用全子集回归来构建预估模型,并采用留一交叉验证进行模型检验。结果表明,构建模型的判定系数R2adj为0.5071,留一交叉验证结果R2cv=0.4860,模型的残差呈均匀的分布在0附近,没有明显的变化趋势。此外,SW检验和NCV检验结果显示残差的正态性和等方差性,表明构建的森林生态功能指数遥感预估模型能够预估森林生态功能指数,为生态功能的快速、经济和定量的评价提供数据支持,为有效森林管理以及决策的制定提供理论支持。
In the present study,a remote sensing estimation model for forest ecological function index is proposed to provide a basis for the evaluation of forest ecosystem in western Hunan.Based on SPOT 5 remote sensing images in western Hunan,the texture and spectral information of field plots were extracted.The all-subsets regression was performed to build the predictive model by including the statistically significant image-derived measures as independent variables and the produced model was further validated for its performance using the leave-one-out cross-validation approach.The results indicated that the adjusted coefficient of determination R 2 adj was 0.507 1,the leave-one-out cross-validation approach R 2 cv was 0.486 0,any particular patterns or trends were not observed from the residual plot of the model.In addition,the results of SW and NCV tests demonstrated the normal distribution.It was concluded that the produced model could predict forest ecological function index,which could provide data support for the rapid,economic and quantitative evaluation of ecological function,and to provide theoretical support for effective forest management and decision-making.