Abstract: Examining crop leaf area index (LAI) and its dynamic change is of great significance for crop growth monitoring and yield estimation. The inversion of crop growth parameters based on terrestrial hyperspectral data is a hotspot in agricultural remote sensing research. However, most previous studies used terrestrial hyperspectral data to establish the estimation model of crop LAI, which is difficult for regional application. In order to apply the terrestrial hyperspectral research results to the satellite scale and realize the inversion of regional peanut LAI, thereby to monitor the largearea peanut growth, we constructed a variety of wideband spectral indices based on the terrestrial observation spectrum data, and built a remote sensing peanut LAI estimation model based on the hyperspectral index. The hyperspectral index was constructed with the GF-1 satellite sensor spectral response function and the terrestrial hyperspectral data on the basis of plot test and field test of the observation station. By comparing the determination coefficient and verification accuracy of different estimation models, we found that the model based on the RVI index (LAI=0.481RVI^0.830) was the best one for LAI estimation. Based on the optimal model, the peanut LAI distribution was obtained by performing the peanut LAI remote sensing mapping, and the remote sensing inversion LAI accuracy was verified by field data. The results showed that the use of wideband index and GF-1 would be suitable for peanut LAI estimation, which has great significance for monitoring large-area peanut growth.

Key words: spectral response function, GF-1, leaf area index, wide band index.