Multi-source Remote Sensing Data Leaf Area Index Inversion

Recently, inversion and validation of Vegetation index and leaf area index (LAI)is one of the hot research topics, and play an important role in the agriculture, animalhusbandry and study and application of ecological environment.This article takesChengDuo County in Qinghai province as the research area, using modern satelliteremote sensing technology, combining GPS measurement technology with LAI-2000plant Crown layer analysis instrument measures its vegetation LAI value and getsfield investigation sampling data. It uses three kinds of satellite data-MODIS, TM andthe environment satellite image to determine the five vegetation index. It establisheslinear regression model and nonlinear regression model between vegetation index andLAI, and determines the statistical correlation coefficient in the model, chooses thebest model in version LAI. In this paper, y I get the following conclusions bycomparing analysis and stud:1The index in the selected five planting, normalized difference vegetation index(NDVI), the ratio vegetation index (RVI), difference vegetation index (DVI), soilvegetation index (SAVI) and transformational vegetation index (TVI), in addition tothe DVI, the regression model established by other index has good correlation.DVI is not suitable in this region for the inversion.2In the LAI-VIS linear regression models,the Modis data and environmentalsatellite NDVI and LAI coefficient of determination R2values are biggest, and havethe highest correlation to TVI; Determination of TM image data TVI and LAIcoefficient R2value are largest, and the highest correlation, so in this study NDVI andTVI is more suitable for modeling and LAI for remote sensing inversion analysis.3Polynomial model in the LAI-VIS nonlinear regression models, the polynomialmodel of three kinds of image data model of the inversion results are best able torespond because LAI is the polynomial model of optimal non-linear inversion model,followed by the power function regression model. so the polynomial model can betterreflect the LAI-VI of the relationship.4In linear and nonlinear regression models established by vegetation index andLAI, nonlinear regression model is superior to linear regression model, it can betterreflect the actual situation of LAI;5In the analog inversion of three remote sensing data, the MODIS data fits bestand is more suitable for inversing with vegetation index and LAI.