• Laser & Optoelectronics Progress
  • Vol. 59, Issue 16, 1630004 (2022)
Jiarui Li1, Jifen Wang1、*, Linyuan Fan1, and Xuejun Shi2
Author Affiliations
  • 1School of Investigation, People’s Public Security University of China, Beijing 100038, China
  • 2Forensic Expertise Center of Beijing Customs Anti-Smuggling Bureau, Beijing 100000, China
  • show less
    DOI: 10.3788/LOP202259.1630004 Cite this Article Set citation alerts
    Jiarui Li, Jifen Wang, Linyuan Fan, Xuejun Shi. Rapid Identification and Classification of Cannabis Oil Based on Data Fusion of Spectroscopy and Chromatography[J]. Laser & Optoelectronics Progress, 2022, 59(16): 1630004 Copy Citation Text show less
    Infrared spectra of cannabis oil
    Fig. 1. Infrared spectra of cannabis oil
    Infrared spectra of all samples
    Fig. 2. Infrared spectra of all samples
    FDA confusion matrix. (a) Single model; (b) fusion model
    Fig. 3. FDA confusion matrix. (a) Single model; (b) fusion model
    Spatial classification detail of cannabis oil
    Fig. 4. Spatial classification detail of cannabis oil
    Eigenvalue importance
    Fig. 5. Eigenvalue importance
    K value selection error rate
    Fig. 6. K value selection error rate
    KNN confusion matrix. (a) Single model; (b) fusion model
    Fig. 7. KNN confusion matrix. (a) Single model; (b) fusion model
    NameIngredient
    IMMUNO BOOSTERMCT Oil,Echinacea Extract,CBD,THC
    NRG BOOSTERMCT Oil,Ashwagandha,Maca Root,Citral-Terpene Blend,CBD,THC
    RELAXMCT Oil,Dang Quai Extract,Red Clover Extract,CBD,THC
    WELLNESSMCT Oil,Passion Flower,Myrcene-Terpene Blend,CBD,THC
    Table 1. Information of cannabis oil samples with different species
    TypeData setSensitivitySpecificityPrecisionAccuracy
    FTIRIMMUNO BOOSTER0.8090.9120.8530.818
    NRG BOOSTER0.8090.9120.853
    RELAX0.8560.6690.729
    WELLNESS0.8330.8620.836
    FTIR+GCIMMUNO BOOSTER0.9831.0000.9920.987
    NRG BOOSTER0.9831.0000.992
    RELAX0.9831.0000.992
    WELLNESS1.0000.9510.974
    Table 2. Parameter results of FDA model
    TypeParameterTraining setTest set
    IMMUNO BOOSTERNRG BOOSTERRELAXWELLNESSIMMUNO BOOSTERNRG BOOSTERRELAXWELLNESS
    FTIRSensitivity0.8220.8070.8220.8110.8400.8240.8490.849
    Specificity0.7580.9470.7640.8390.8540.9320.7890.789
    Precision0.7860.8640.7900.8240.8470.8710.8160.816
    Accuracy0.7310.9570.7410.8460.8570.9410.7690.842
    FTIR+GCSensitivity1.0001.0001.0001.0001.0001.0001.0001.000
    Specificity1.0001.0001.0001.0001.0001.0001.0001.000
    Precision1.0001.0001.0001.0001.0001.0001.0001.000
    Accuracy1.0001.0001.0001.0001.0001.0001.0001.000
    Table 3. Parameter results of KNN model
    Jiarui Li, Jifen Wang, Linyuan Fan, Xuejun Shi. Rapid Identification and Classification of Cannabis Oil Based on Data Fusion of Spectroscopy and Chromatography[J]. Laser & Optoelectronics Progress, 2022, 59(16): 1630004
    Download Citation