In order to establish a scientific method for inspecting residual cigarette liners on site, the latest differential spectroscopy technology was used to inspect 44 different brands and different series of samples, and the samples were classified according to the different main fillers in the samples. Combined with chemometrics, the data reduced by principal component analysis was used for cluster analysis, and the significant P value and Pearson correlation coefficient were used to evaluate the quality of the clustering results. The optimal clustering results were used to establish discrimination formula. The results show that the samples can be divided into 4 categories after analyzing and comparing the spectra. The established discriminant realized 100% accurate classification of 44 samples, and provided a basis for the discrimination of unknown samples. In summary, the method combined with spectral analysis and chemometrics can quickly, non-destructively and accurately make objective inspections of samples. It has certain universality and provides a theoretical basis for the actual handling of cases by police.