A rapid and non-destructive identification method for shiitake geographical origins was developed based on near infrared(NIR) spectroscopy combined with chemometrics.The NIR spectra of dried shiitake samples from different regions were collected in the diffuse reflectance mode.The origin discrimination models for shiitake cultivated in Jilin,Hubei and Fujian provinces were respectively established by partial least squares discriminant analysis(PLSDA),then optimized by spectral preprocessing and wavelength screening technique,and finally verified with samples.Results showed that the model established with the original spectrum could initially achieve the identification of the origin.The spectral preprocessing was used to subtract the background information in the spectrum,and the wavelength selection technique was used to select a specific wavelength to optimize the model.The prediction results could be further improved based on spectral preprocessing and wavelength selection.This method provides a new approach for tracing the origin of shiitake production,which has an important practical significance in the development of shiitake industry.