• Spectroscopy and Spectral Analysis
  • Vol. 42, Issue 11, 3501 (2022)
Near Infrared microplastic measuring system1: Optical fiber; 2: Sample; 3: Integrating sphere with light source; 4: Receiving optical fiber; 5: Spectral collection module; 6: Spectral processing module; 7: Photoelectric conversion module; 8: Date processing module
Fig. 1. Near Infrared microplastic measuring system
1: Optical fiber; 2: Sample; 3: Integrating sphere with light source; 4: Receiving optical fiber; 5: Spectral collection module; 6: Spectral processing module; 7: Photoelectric conversion module; 8: Date processing module
K-fold cross-validation structure
Fig. 2. K-fold cross-validation structure
XGBoost and SVM for microplastic detection
Fig. 3. XGBoost and SVM for microplastic detection
Original Near Infrared spectra of microplastics
Fig. 4. Original Near Infrared spectra of microplastics
The 30 most important spectral variables
Fig. 5. The 30 most important spectral variables
CV average accuracy of n_estimators, learning_rate, max_depth, gamma
Fig. 6. CV average accuracy of n_estimators, learning_rate, max_depth, gamma
SVM and XGBoost confusion matrixes
Fig. 7. SVM and XGBoost confusion matrixes
Accuracy score, precision score, recall and F1-score of 10-fold cross validations of XGBoost and SVM
Fig. 8. Accuracy score, precision score, recall and F1-score of 10-fold cross validations of XGBoost and SVM
Microplastic
category
Precision of
XGBoost
Recall of
XGBoost
F1-score
of XGBoost
Precision
of SVM
Recall of
SVM
F1-score
of SVM
ABS1.001.001.001.001.001.00
EVA1.001.001.001.001.001.00
PAN0.931.000.970.821.000.90
PBT1.001.001.001.000.840.91
PC0.861.000.921.000.890.94
PCL0.911.000.950.951.000.98
PES1.001.001.001.001.001.00
PET1.000.870.930.851.000.92
PLA0.951.000.981.000.760.86
PMMA1.000.940.970.941.000.97
POM1.000.830.900.880.960.92
PP0.951.000.970.901.000.95
PPO1.001.001.000.840.890.86
PPS1.001.001.001.001.001.00
PS0.900.900.900.950.900.93
PTFE1.000.950.971.000.950.97
PVA1.001.001.001.000.950.97
PVC0.941.000.970.941.000.97
SBS1.001.001.000.951.000.98
TPU1.001.001.001.000.810.90
Table 1. Comparison of operation results between XGBoost and SVM algorithms