Table 5 Results of the classification of the segments of the live larvae into head, normal and abdomen end.
From: Multipurpose monitoring system for edible insect breeding based on machine learning
Segment type | Model | \(TPR_{seg}\) | \(PPV_{seg}\) | \(F1_{seg}\) | |||
|---|---|---|---|---|---|---|---|
Raw predictions | Processed predictions | Raw predictions | Processed predictions | Raw predictions | Processed predictions | ||
Normal | LDA | 0.895 | 0.983 | 0.988 | 0.989 | 0.939 | 0.986 |
LogReg | 0.930 | 0.990 | 0.996 | 0.996 | 0.962 | 0.993 | |
SVM | 0.920 | 0.986 | 1.000 | 1.000 | 0.958 | 0.993 | |
Head | LDA | 0.833 | 0.833 | 0.600 | 0.769 | 0.698 | 0.800 |
LogReg | 0.750 | 0.750 | 0.692 | 0.794 | 0.720 | 0.771 | |
SVM | 0.750 | 0.750 | 0.692 | 0.794 | 0.720 | 0.771 | |
Abdomen end | LDA | 0.750 | 0.750 | 0.551 | 0.771 | 0.635 | 0.760 |
LogReg | 0.806 | 0.806 | 0.558 | 0.725 | 0.659 | 0.763 | |
SVM | 0.833 | 0.833 | 0.536 | 0.714 | 0.652 | 0.769 | |
All (average metrics) | LDA | 0.826 | 0.855 | 0.713 | 0.843 | 0.757 | 0.849 |
LogReg | 0.829 | 0.849 | 0.749 | 0.838 | 0.780 | 0.842 | |
SVM | 0.834 | 0.856 | 0.743 | 0.836 | 0.777 | 0.844 | |
