 The study reconstructs time series vegetation index, 6, using normalized difference vegetation index, NDVI, an enhanced vegetation index, EVI, from Landsat underscore 5 underscore TM and Landsat underscore 7 underscore ETM, an estimate crop yield based on the reconstructed model. The results show that EVI is better for reconstruction of time series 6 curves than NDVI, with a correlation coefficient of 0.90 during the growth stage. The extreme model reconstruction reduces background differences between different plots, providing the basis for regional yield estimation using remote sensing. With increases in less than 30M resolution observation programs, this model is expected to be increasingly used, improving the applicability of VI-based yield estimation models. This article was authored by Ling He-Meng, Shin Le-Jong, Wanzhu and Liu, and others.