Table 3 Performances of the proposed model in terms of different evaluation metrics (in %) on the “JUHCCR-v1.o” dataset across five runs of the training process.
From: JUHCCR-v1: a database for hand-drawn electrical and electronics circuit component recognition
Run | Accuracy | Precision | Recall | F1-score |
|---|---|---|---|---|
1 | 91.15 | 91.87 | 91.15 | 91.51 |
2 | 91.28 | 91.95 | 91.38 | 91.66 |
3 | 91.02 | 91.72 | 90.96 | 91.34 |
4 | 90.89 | 92.03 | 91.08 | 91.55 |
5 | 91.31 | 91.84 | 91.28 | 91.56 |
Overall | 91.13 ± 0.16 | 91.88 ± 0.10 | 91.17 ± 0.15 | 91.52 ± 0.11 |
