Simulation And Analysis Of Tree Canopy BRF Based On The Real Scene

Early in the remote sensing application,people often have assumed that the objects on the Earth's surface are Lambert reflectors which have no relation or connection with the incident radiation direction or the observation direction in the study of the reflectivity.Although this assumption has been used simply and conveniently in the process of modeling or calculation,a lot of observations has proved that all the objects on the Earth's surface are non-Lambertian in some ways,and have bi-directional reflectance characteristic,especially in vegetation reflectance.With the development of remote sensing technology including the application of the multi-angle remote sensing satellite and the development of the theoretical research for bi-directional reflectance distribution.There are many of models for characterizing bi-directional reflectance such as Radiative Transfer model,Geometrical Optics model,GORT model and Computer Simulation model and they have become the hot field in remote sensing science to calculate vegetation canopy bi-directional reflectance and have been widely recognized in the world.In the field of vegetation remote sensing,BRDF(Bi-directional Reflectance Distribution Function)has always been an essential theoretical foundation for multi-angle vegetation remote sensing.It's difficult to obtain a given accuracy because of the complexity of vegetation structure,the error from experimental observation,the limitation of the satellite data and the model of reflection characteristics of an adaptive resistance.Particularly for the larger trees,there is more difficult to calculate accurately its BRF(Bi-directional Reflectance Factor)according to the real structure of trees.Given this the paper has simulated and analyzed the BRF of tree canopy using Radiosity method based on a 3-D real scene.In this study,Radiosity method is used to simulate the canopy BRF of vegetation,which is one of the methods in computer simulation model category.The Radiosity method composes of two modules: 1)using L-system to render the 3-D realistic scenes;2)based on the rendering scenes,simulating the canopy bi-directional reflectance using Radiosity method.In this paper,the major research contents are as following:The writer simulates the BRF of single-tree at solar principal plane and cross solar principal plane using Radiosity-Graphics combined Model(RGM).Its input files include three-dimensional structure file,spectral parameters of all components,the Sun and sky definition file,the temperature field,soil file,etc.,and computes the tree canopy BRF at the visible light and near infrared band.In order to research how the canopy BRF is changing along with varying some model parameters,the writer simulates the canopy BRF at each band under the different conditions of tree components,leaf structure,solar position,scene parameters including field of view,sky light ratio,and underlying surface type,and then analyzes the corresponding distribution respectively.The results show that the canopy BRF is changed regularly as the parameter changes,such as the canopy BRF value would decreases on the contrary in case of leaf area index increased,the hotspot point of canopy BRF at each band would change with solar zenith angle,hotspot effect of the canopy BRF would be weakened by field of view increased,the more diffuses light,the less BRF,and the canopy BRF would be associated with changes in the different types of underlying surface.Based on the above conclusions and combined with the measured data,the writer simulates 3-D real scene by means of extension L-system and calculated canopy BRF of the whole scene,then analyzes the correlation between computer simulation BRF and measured BRF.The results show that the mean correlation coefficients between all computer simulation BRFs and measured BRFs are greater than 0.7,indicating the simulated canopy BRF using RGM model is relatively accurate and the field experiment in this paper also is proved to be reliable.