The Processing of traditional Chinese medicine (TCM) is the key to clinical application of TCM, and processing has functions such as enhancing the efficacy, attenuating the toxicity andmoderating medicine property. In order to the realizing safe, reasonable and effective use of medicine in clinical, research on identification of TCM processed products is of great significance. The Gentiana rigescens samples which processed with five different methods were discriminated by Fourier transform infrared spectroscopy (FTIR). Baseline correction and normalization were used to pretreat all original spectra and the noise was cut off. The spectra range was from 3 400 to 600 cm-1. The effect of multiple scattering correction and standard normal variable on the model were observed and compared. Samples were divided into calibration set and prediction set at the ratio of 3∶1. The principal component analysis (PCA) was applied to reduce data dimensionality and discriminant analysis model was established. The result indicated that the main absorption peaks of samples were 3 378, 2 922, 1 732, 1 610, 1 417, 1 366, 1 316, 1 271, 1 068, 1 048 cm-1 which 1 738, 1 643, 1 613, 1 420, 1 051 cm-1 as to gentiopicrin; 1 068, 1 048, 935 cm-1 as to carbohydrate. The accumulation contribution rate of first three principal components is 94.05%. Most of the information reflected the original data. There were differences among different samples. The result of discriminant analysis showed that the recognition rate of G. rigescens samples could achieve to 100% based on baseline correction and normalization treatment combined with MSC with the precondition of principal component scores being 10. In conclusion, FTIR is a feasible, rapid and non-destructive method to discriminate G. rigescens samples wtih different processing methods. It also provided reference for discrimination of processed products of medicine materials.