Abstract: A new test and classification method for disposable masks was explored.Disposable masks produced by 31 different manufacturers were collected and tested with differential Raman spectrometer,and the differential Raman spectra were obtained.The difference Raman spectra were compared and divided into 4 categories manually.IBM SPSS Statistics 25.0 software was used to standardize the data,and then systematic cluster analysis and K-means cluster analysis were used to classify the data,and finally Fisher discriminant analysis was used to verify the above classification methods and classification results.The results showed that when the parallel distance was the lowest,disposable mask samples could be divided into 6 categories,which basically further subdivided the samples on the basis of manual classification.This method,which combines differential Raman spectroscopy with stoichiometry,has the advantages of non-destructive and efficient,and can provide help for the forensic scientific and technical examination and identification of material evidence.