A prediction discrimination model of fresh and stale rice was established based on Raman Spectroscopy and Chemometrics. The pretreatment processes have to be employed before the experiment. A total of 60 samples were put in the special box. The samples were measured by 785nm Raman spectrometer, which can collect spectral range of 200~2 400 cm-1. Smoothing, baseline correction were conducted to process the raw spectra. Principal Component Analysis (PCA) was employed to reduce dimension analysis of full-wave band of fresh and stale rice, and it could classify the samples preliminarily. The discrimination model was developed with Partal Least Squares (PLS). The correct classification rates in the training set and prediction set were 100% and 95%, respectively. The results in this research indicated it is a quickly useful method to discriminate between fresh and stale rice.