The feasibility of using the Fourier transform near-infrared diffuse reflectance spectrometry to determine egg quality and storage time rapidly was investigated. The spectral models were established based on Partial Least Square(PLS) for egg quality indexes(Haugh unit, protein pH and egg shape) and storage time and the coefficient of determination(R2cv) and the Root Mean Square Error of Cross Validation(RMSECV) were used to evaluate the model accuracy. The results show that the optimal predictive models for Haugh unit, protein pH, egg shape and storage time are obtained by effective information selection with the R2cv of 0.86、0.84、0.26 and 0.92, respectively and the RMSECV of 7.52, 0.17, 0.05 and 1.37, respectively. It concludes that the models have better measuring accuracy and prediction capabilities except for the egg shape, which can meet the requirements of fast non-destructive detection for the fresh eggs. Moreover, All parameters have a higher correlation with the change of storage time except for the egg shape, which meets the change tendency of the storage fresh eggs.