Leaf area index (LAI) is an important indicator of wetland ecosystem health. Poyang Lake wetland vegetations grow densely, with LAI of large dynamic range. Considering the complex scattering mechanisms of radar data, a radar vegetation index was defined. To overcome the saturation of the optical vegetation indices, a new integrated vegetation index using GF-1 and Radarsat-2 data was established for estimation of wetland vegetation LAI. The validation of measured data and theoretical model simulation showed that this integrated vegetation index is a good alternative to that using only the optical or radar observation. The best fitting models were built with optical vegetation indices, radar vegetation index, and the integrated vegetation index, respectively. The result indicates that the integrated vegetation index can improve predication accuracy for wetland vegetation LAI.