Abstract: Early detection of wheat grain protein content using remote sensing technology is very helpful to optimize field management. An agricultural parameter, Nitrogen Nutrition Index (NNI), was introduced in this study. It can be used as an intermediate variable between remote sensing data and grain protein content, which would enhance accuracy of remote detected wheat grain protein content. Field campaign was conducted to obtain remote sensing data and agricultural parameters at flag growth stage of winter wheat. Using these data, the abilities of nitrogen nutrition index and other agricultural parameters for grain protein content prediction were compared. Then, Principal Component Regression method was used to establish grain protein content prediction model based on relationships between remote sensing data and NNI, and between NNI and grain protein content. The results showed NNI was the best agricultural parameter for grain protein content detection, compared with other agricultural parameters. The established prediction model, using NNI and remote sensing data, can detect wheat grain protein content accurately, with a R2 value of 0.48, a root-mean-square-error (RMSE) value of 0.38%, and a relative error value of 2.32%.