LAI Inversion Method For Crop Based On LiDAR And Multispectral Remote Sensing

Datian planting is the main planting method of crops in Xinjiang.Monitoring of crop condition monitoring is the monitoring of the whole growth cycle of crops and their changes.It is an important basis for reflecting agricultural information,guiding production,and macro management decisions.As a parameter describing the canopy structure of crops,leaf area index is not only an important parameter to characterize crop canopy structure and determine crop growth,but also a key factor in determining biomass and yield.In the fine agricultural remote sensing monitoring,how to quickly obtain the crop leaf area index,it is very important to judge the crop growth.At present,the use of remote sensing technology to carry out large-scale resource surveys and remote sensing monitoring of crops is relatively mature from a technical point of view,but there are few studies using new sensors such as high-precision laser radar to invert the leaf area index.The research work of this paper is to use the LIDAR airborne technology of UAV to realize the inversion of crop LAI by using multi-spectral remote sensing technology.The main contents,results and conclusions are:1.Flight planning and data acquisition programs based on different flight platforms.Utilizing the characteristics of fast and highly efficient drones,the appropriate research area was selected to obtain laser radar and multispectral data.Solved the problem of coverage calculation,flight route design,ground target laying,and data matching in the data acquisition process,and laid the foundation for obtaining effective data.2.Accuracy evaluation of airborne LiDAR crop point cloud data without calibration points.Based on the analysis of the point cloud acquisition and positioning model of airborne laser radar,the laser ranging errors and dynamic delay errors are quantitatively described and analyzed from two aspects:system error and random error.Using the spatial topological relationship of the point cloud points,the true value of the fitted elevation model and the fitting equation of the plane model are obtained.Based on this,the point cloud data accuracy without correction point is evaluated.The experimental results show that the maximum residual error of height accuracy is 5.6cm,and the maximum residual value of plane accuracy is 2.78cm,which verifies the validity of the accuracy of crop canopy parameter extraction data.3.LAI inversion method for airborne LIDAR crops.High-density point cloud data was obtained based on small light spot LiDAR,ground points and non-ground points were separated,and the digital surface model（DSM）and digital elevation model（DEM）of the cotton fields in the study area were obtained through pre-processed cotton high-density point cloud data.The difference operator obtains its canopy height model（CHM）,and then extracts effective canopy structure parameters.The LAI inversion model of cotton was constructed by using correlation analysis method to select laser penetration index（LPI）,echo point cloud density（D）,porosity(f_gap),and normalized elevation value(V_nDSM)with correlation coefficient greater than 0.2.Accuracy verification and evaluation are performed with the measured leaf area index.The experimental results show that the coefficient of determination between the model’s LAI and the measured LAI is 0.824,and the root mean square error is 0.072.4.Crop LAI inversion based on multi-source remote sensing data.Using Lidar and multi-spectral sensors to obtain vertical and horizontal structure parameters of crops at the same time,spectral parameters with higher correlation with measured LAI were obtained by stepwise regression method:optimized chlorophyll absorption rate index(MCARI₆₅₆),and improved chlorophyll absorption rate index（TCARI）And LiDAR canopy parameters:laser penetration index（LPI）,echo point cloud density（D）,porosity(f_gap)and other parameters for leaf area index inversion,estimated LAI value and measured LAI value determination coefficient is 0.85.Then the model of LAI inversion of crop LAI was optimized.