Abstract: The results are achieved through the canopy hyperspectral study in the different growing stages and the different varieties of cotton. The spectral difference of the visional wave is obvious in 2.5 of LAI, and the difference trend to the saturation ; the correlation between LAI and the spectrum is significant negative in the visional wave, and the correlation between LAI and the near infrared is significant positive; there are three high plaint regions, including blue region(523-531nm)、yellow region(570-576nm)、red region(700-755nm) respectively, resulted from the correlation between LAI and the derivative spectrum, and among them the red region is the strongest. The late peak of the red edge is the leaf area index. Among hyperspectral parameters constructing the estimation model of LAI of cotton, the analyses shows that the relative error, including NDVI(660、800)、NDVI(550、800)、NDVI(500、800)、NDVI(670、800)、SDy(570-573nm)、SDr(714-755nm)、D723、Dr, is under 30% and RMSE is under 0.6, and NDVI(660、800)and NDVI(550、800) is better comparatively. The achievement is improved throughout establishing the stage models, and SDr is used when the leaf index is under 1.0 and SDr/SDb is used when the leaf area index is above 1.0.

Key words: cotton, the leaf area index , the hyperspectral response, the inversion of remote sensing