Environmental stress of light and temperature is a major restricting factor that affects the quality and yield of crops. Traditional crop stress monitoring is insufficiently sensitive, time-consuming and laborious, and mostly destructive testing. In recent years, with the rapid development of information technology, hyperspectral technology can quickly and non-destructively obtain crop physiological information, and dynamically monitor the response to adversity, providing digital support for the precision production and intelligent decision-making of modern agriculture, and is of great significance for realizing the transformation of traditional agriculture to precision and modern digital agriculture. This paper takes the corn seedling stage as the research object, obtains the hyperspectral data and physiological parameters of leaves under different light and temperature environments, explores the response law of corn leaves to different light and temperature environments, conducts hyperspectral difference analysis, and construct physiological parameters Hyperspectral inversion model. The correlation analysis method is used to screen the spectral sensitive band. The preprocessing method combining Multivariate Scattering Correction (MSC), Standard Normal Variable transformation (SNV), and Savitzky-Golay (SG) smoothing is used, respectively. Partial Least Square regression (PLS), Principal Component Regression (PCR), Stepwise Multiple Linear Regression (SMLR) three modeling methods combination, the model correlation coefficient and root mean square error are used as model effect evaluation indicators to explore the optimal method of hyperspectral inversion of leaf physiological parameter models. The results show that the hyperspectral characteristics of corn under different light and temperature environments have the same changing trend as a whole, but there are still differences. The reflectance of the spectrum in the 500~700 nm band gradually increases with the increase of light intensity, the reflectivity of the spectrum in the 760~900 nm band gradually increases with the increase of temperature, and the changes of the light and temperature stress environment can be reflected in the hyperspectral characteristics. The spectral reflectance in the 760~900 nm band is relatively high in a high temperature stress environment, the spectral reflectance is low in a low light stress environment, and the reflectance is significantly reduced in a low temperature stress environment. The optimal combination of SPAD and F_v/F_m inversion model is PLS-MSC-SG, the correlation coefficients of the model validation set are 0.958 and 0.976, and the correlation coefficients of the training set are 0.979 and 0.995, respectively. The model’s predictive accuracy is high, which indicates that the use of hyperspectral technology can realize quantitative monitoring of maize plants under light and temperature environmental stresses, improve the level of refined management in the field, and provide a reference for the intelligent management of high-quality and high-yield maize.