Abstract: Leaf area index (LAI) is one of the most important biophysical parameters of crop and other land surface vegetation. In order to improve the accuracy of retrieving leaf area index of winter wheat using HJ CCD data, the accuracy of different vegetation indices and regression models were compared and analyzed from the aspects of growth stages on the basis of five common vegetation indices including NDVI (normalized difference vegetation index), EVI (enhanced vegetation index), EVI2 (two-bands enhanced vegetation index), RVI (ratio vegetation index) and SAVI (soil adjusted vegetation index) and three common regression models. The results showed that LAI and all five vegetation indices had good correlative relationship except at reproductive growth stage. Exponential and linear regression model were the best regression models for the whole growth stage and the vegetative stage, respectively. EVI performed better than other four indices when simulated at the whole growth stage (R2=0.9348), and SAVI was the best index for LAI retrieving at vegetative growth stage (R2=0.9404). This paper provides the reference for LAI inversion by vegetation index of winter wheat.