Freeze injury, one of the most destructive agricultural disasters caused by climate, has a significant impact on the growth and production of winter wheat. Chlorophyll content is an important indicator of a plant's growth status. In this study, we analyzed the hyperspectral reflectance of normal and freeze-stressed leaves of winter wheat using a spectro-radiometer in a laboratory. The response of the chlorophyll spectra of plants under freeze stress was analyzed to predict the severity of freeze injury. A continuous wavelet transform (CWT) was conducted in conjunction with a correlation analysis, which generated a correlation scalogram that summarized the correlation between the chlorophyll content (SPAD value) and wavelet power at different wavelengths and decomposition scales. A linear regression model was established to relate the SPAD values and wavelet power coefficients. The results indicated that the most sensitive wavelet feature (region E: 553 nm, scale 5, R2 = 0.8332) was located near the strong pigment absorption bands, and the model based on this feature could estimate the SPAD value with a high coefficient of determination (R2 = 0.7444, RMSE = 7.359). The data revealed that the chlorophyll content of leaves under different low temperatures treatments could be accurately estimated using CWT. Also, this emerging spectral analytical approach can be applied to other complex datasets, including a broad range of species, and may be adapted to estimate basic leaf biochemical elements, such as nitrogen, cellulose, and lignin.