基于NDVI与偏最小二乘回归的荒漠化地区植被覆盖度高光谱遥感估测

摘要/Abstract

摘要： 以星载高光谱影像Hyperion为数据源,系统比较了NDVI与偏最小二乘回归（PLS）估测荒漠化地区植被覆盖度的能力,模型的建立（n=46）与独立检验所用样本（n=10）均为地面实测数据。研究结果表明,基于星载高光谱数据的NDVI与PLS模型可以有效地估测荒漠化地区植被覆盖度。相比于宽波段NDVI（RMSEP=10.5618）及基于803.3/671.02 nm计算的标准高光谱NDVI（RMSEP=8.3863）,选择特定高光谱波段（823.65/701.55 nm）构建的NDVI预测植被覆盖度的误差明显较低（RMSEP=6.5189）。基于高光谱所有波段原始反射率、一阶导数及包络线去除光谱的PLS回归模型表现,要明显优于仅利用两个波段信息的NDVI,其中基于原始反射率的PLS回归模型表现最佳,RMSEP为4.4998,约为因变量平均值的23%。

关键词: Hyperion, NDVI, 偏最小二乘回归, 植被覆盖度, 一阶导数, 包络线去除

Abstract: Based on Hyperion data, NDVI and partial least squares regression (PLS) are used to estimate the vegetation cover in desertified region. The predictive accuracy of NDVI and PLS regression models are then determined and compared using calibration (n=46) and test (n=10) data. The results show that both NDVI and PLS regression models can estimate vegetation cover effectively. Compared with broadband NDVI (RMSEP=10.5618) and standard hyperspectral NDVI involving bands at 803.3 nm and 671.02 nm (RMSEP=8.3863), the NDVI computed from bands at 823.65 nm and 701.55 nm has a lower standard error of prediction (RMSEP=6.5189). PLS regression models based on original, derivative and continuum-removed spectra, which utilize all bands of Hyperion data, produce lower prediction errors (RMSEP is between 4.4998~5.1449) than NDVIs, which justly use two bands of Hyperion data. The lowest prediction error (RMSEP=4.4998, 23% of mean) is obtained by PLS regression analysis involving original reflectance.

Key words: Hyperion, NDVI, partial least squares regression, vegetation cover, first derivative, continuum-removed

中图分类号:

李晓松;李增元;高志海*;白黎娜;王琫瑜. 基于NDVI与偏最小二乘回归的荒漠化地区植被覆盖度高光谱遥感估测[J]. 中国沙漠, 2011, 31(1): 162-167.

LI Xiao-song;LI Zeng-yuan;GAO Zhi-hai;BAI Li-na;WANG Beng-yu . Estimation of Vegetation Cover in Desertified Regions from Hyperion Imageries Using NDVI and Partial Least Squares Regression[J]. JOURNAL OF DESERT RESEARCH, 2011, 31(1): 162-167.

参考文献

[1]宋怡,马明国.基于SPOT VEGETATION数据的中国西北植被覆盖变化分析[J].中国沙漠,2007,27(1):89-93.
[2]祁元,王一谋,王建华,等.基于遥感和GIS的荒漠化动态分析——以宁夏盐池为例[J].中国沙漠,2003,23(3):275-279.
[3]丁建丽,塔西甫拉提·特依拜,刘传胜.策勒绿洲植被覆盖动态变化遥感研究[J].中国沙漠,2003,23(1):79-83.
[4]陈晓光,李剑萍,李志军,等.青海湖地区植被覆盖及其与气温降水变化的关系[J].中国沙漠,2007,27(5):797-801.
[5]高志海,李增元,魏怀东,等.干旱地区植被指数(IV)的适宜性研究[J].中国沙漠,2006,26(2):243-248.
[6]Huete A R,Jackson R D.Soil and atmosphere influences on the spectra of partial canopies [J].Remote Sensing of Environment,1988,25(1):89-105.
[7]Qi J,Moran M S,Cabot F,et al.Normalization of sun/view angle effects using spectral albedo-based vegetation indices[J].Remote Sensing of Environment,1995,52(3):207-217.
[8]Thenkabail P S,Smith R B,De Pauw E,et al.Hyperspectral vegetation indices and their relationships with agricultural crop characteristics[J].Remote Sensing of Environment,2000,71(2):158-182.
[9]Gong P,Pu R L,Biging G S,et al.Estimation of forest leaf area index using vegetation indices derived from Hyperion hyperspectral Data[J].IEEE Transactions on Geoscience and Remote Sensing,2003,41(6):1355-1362.
[10]Lee K S,Cohen W B,Kennedy R E,et al.Hyperspectral versus multispectral data for estimating leaf area index in four different biomes[J].Remote Sensing of Environment,2004,91(3-4):508-520.
[11]Huang Z,Turne B J,Dury S J,et al.Estimating foliage nitrogen concentration from HYMAP data using continuum removal analysis[J].Remote Sensing of Environment,2004,93(1-2):18-29.
[12]USGS.EO1 products introduction[EB/OL].(2006-01-01)[2008-05-15].http://eo1.usgs.gov/products.php.
[13]谭炳香,李增元,陈尔学,等.EO-1 Hyperion高光谱数据的预处理[J].遥感信息,2005(6):38-43.
[14]Rondeaux G,Steven M,Baret F.Optimization of soil-adjusted vegetation indices[J].Remote Sensing of Environment,1996,55(2):95-107.
[15]Cloutis E.Hyperspectral geological remote sensing: evaluation analytical techniques[J].International Journal of Remote Sensing,1996,17(12):2215-2242.
[16]Hansen P M,Schjoerring J K.Reflectance measurement of canopy biomass and nitrogen status in wheat crops using normalized difference vegetation indices and partial least squares regression[J].Remote Sensing of Environment,2003,86(4):542-553.
[17]Cho M A,Skidmore A,Corsi F,et al.Estimation of green grass/herb biomass from airborne hyperspectral imagery using spectral indices and partial least squares regression[J].International Journal of Applied Earth Observation and Geoinformation,2007,9(4):414-424.