Background Radix Astragali is a medicinal herb with various physiological activities and a long history of use dating back 2,000 years. Tens of thousands of tonnes of Radix Astragali are sold worldwide each year. Their clinical applications are affected by many factors, including geographical origin. Conventional microscopic examination and chromatography-based fingerprinting of Radix Astragali roots are tedious extraction procedures, resulting in loss of their original form. Raman spectroscopy is a non-invasive and non-destructive technique that can be used in the in-situ analysis of herbal samples. In this study, we investigated the potential of a 1064 nm-excited portable Raman spectrometer and data fusion for the rapid analysis of Radix Astragali samples from different sources and origins.Results A portable Raman spectrometer was used for the analysis of certified and counterfeit Radix Astragali samples as well as for the determination of their geographical origin. Dispersive Raman scattering, excited at 1064 nm, produced minimal fluorescence background and facilitated easy detection of the weak Raman signal. By moving the Raman probe point-by-point from the centre of the sample to the margin, the spectral fingerprints, composed of dozens of Raman spectra representing the entire Radix Astragali samples, were obtained. Principal component analysis and partial least squares discriminant analysis (PLS-DA) were applied to the Radix Astragali spectral data to compare the classification results, leading to efficient discrimination between genuine and counterfeit Radix Astragali. However, there were high similarities among Radix Astragali samples from different regions owing to the similarities in their main chemical compositions. In the PLS-DA model using data fusion combined with different pre-processing methods, the samples from Shanxi Province were separated from those belonging to other habitats.Conclusion Using a combination of 1064-nm excitation and point-by-point spectral collection mode, the fused Raman spectral data can effectively improve the recognition rate and accuracy of herbal samples, which can be a valuable tool for the identification of genuine medicinal herbs.