Abstract: Leaf area index (LAI) is an important parameter of vegetation ecosystems, which can represent the growth situation of vegetation. The PROBA(project for onboard autonomy)/CHRIS（compact high resolution imaging spectrometer）data acquired in June 4, 2008 was used to inverse LAI of spring wheat combing with the radiative transfer model (PROSAIL) and ANN(artificial neural network), and to validate the results according to the in-situ measurements. The optimal bands were selected using segmented principal component analysis. Three bands(center wavelength 551.1 nm、696.9 nm and 871.5 nm, respectively) were finally used to inversion of LAI. The selected combination of three observation angles (0°, 36° and 55°) shows high accuracy inversion LAI with R2=0.854，RMSE=0.344, MAE=0.213. The accuracy of inversion LAI can be improved with increasing the number of observation angle. However, if the number of angles is more than three, the accuracy will conversely decrease because of the uncertainty augment of multi-angle data.