Study On Detection Of Mineral Mongolian Medicine Based On Raman Spectroscopy

Objective:Raman spectroscopy and machine learning methods are used to classify and identify the easily confused mineral Chinese and Mongolian medicinal materials,providing a simple,rapid and effective new method for the quality evaluation of clinical compound preparations and safe medication level.Methods:Through Raman spectrum technology determination of the sulfur,carbonate,sulfate of the confusing minerals,and find their characteristic peak and belonging,on this basis,build visual identification method based on Raman spectrum technology,and combined with principal component analysis(PCA)-support vector machine(SVM),realize the rapid identification method of spectrogram,and two batches of sulfur(buy different)and two batches of cold stone(buy different)and two batches of nitrate(buy different)classification.Results:Although its Raman spectra are very similar and almost indistinguishable with the naked eye,the PCA-SVM algorithm enables accurate classification and discrimination of the spectrographs with 100% accuracy.Conclusion:This method has the advantages of fast detection,accurate data,non-destructive,convenient,low cost,and high sensitivity to mineral drugs,which can supplement the shortcomings of complex pretreatment process,long detection time and high cost of other detection methods.By detecting several confusing mineral medicinal materials,this method has important application value in the quality evaluation of clinical compound preparations and the level of safe drug use.