Abstract: the use of geometric optical models in the retrieval of FO resting crown properties, spectral reflectance of each component should be referred to as a priori knowledge. In general, these reflectance spectra have been investigated by analytical instrumentation. The purpose of this paper is to obtain forest structural parameters, the use of light detection and ranging (LIDAR) data, and uses a linear spectral unmixing model to determine the spectral reflectance of scene component, which is regarded as the retrieval of forest canopy coverage and plant area index prior knowledge (of) to simplify the Li Strahler geometric optical model based on SPOT-5 - HRG image. He is the first one in the airborne laser radar 3D inversion of forest structure parameters, is not affected by the proportion of pixels of each pixel in the shadow crown or kg of sample and calculated, which is used to the spectral reflectance of scene component decomposition model is extracted by a linear spectrum. Finally, the forest canopy density and the inversion of the geometric optical model. In order to obtain the difference between the SPOT-5 image time and the measured data, it is necessary to verify the results of forest canopy cover about 2 months later. Relatively high values of R2 between retrieval results and measurements show the efficiency of our approach.