Detection method for analyzing xanthium sibiricum adulteration through near infrared spectrum and GBDT algorithm

The invention discloses a method applied to rapid detection of certified traditional Chinese medicine xanthium sibiricum and counterfeit xanthium sibiricum doped with substance highly similar characters. The near infrared spectrum of a sample is collected to scan the spectrum of the sample, first derivation, vector normalization, filtering smoothing and other preprocessing are carried out on spectrogram data, principal component information representing the sample to the maximum degree is extracted, and a GBDT algorithm is used for constructing a prediction model for rapidly analyzing certified xanthium sibiricum, adulterated xanthium mongolicum and adulterated Xanthium inaequilatum. The detection method does not need to use a large number of reagents, is green and environment-friendly, simple and convenient to operate, good in reproducibility and stable and accurate in result, can realize rapid online detection and real-time evaluation feedback when the amount of a xanthium sibiricum sample is large, and is