Table 4. Comparison with State-Of-The-Arts methods (in %).
| Model | Sen | Spc | Prc | Acc | F1 | MCC | FMI |
|---|---|---|---|---|---|---|---|
| WRE+ 3SBBO (Wu 2020) | 86.40 ± 3.00 | 85.81 ± 3.14 | 86.14 ± 3.03 | 86.12 ± 2.75 | 86.16 ± 2.77 | 72.42 ± 5.55 | 86.15 ± 2.76 |
| GoogLeNet-COD-A (Yu, Wang et al. 2020) | 90.54 ±2.16 | 82.77 ±2.65 | 84.07 ±1.93 | 86.66 ±1.14 | 87.15 ±1.06 | 73.59 ±2.25 | 87.23 ±1.07 |
| GLCM-SVM (Chen 2021) | 72.03 ±2.94 | 78.04 ±1.72 | 76.66 ±1.07 | 75.03 ±1.12 | 74.24 ±1.57 | 50.20 ±2.17 | 74.29 ±1.53 |
| 6L-CNN (Hou and Han 2022) | 89.47 ±1.50 | 87.47 ±2.11 | 87.75 ±1.76 | 88.47 ±1.05 | 88.59 ±0.99 | 76.98 ±2.09 | 88.60 ±0.99 |
| SIDCAN (Zhang, Zhang et al. 2021) | 92.86 ±1.59 | 93.64 ±2.09 | 93.36 ±2.02 | 93.26 ±0.74 | 93.08 ±0.71 | 86.55 ±1.49 | 93.10 ±0.72 |
| PZM-DSSAE (Khan 2021) | 92.06 ±1.54 | 92.56 ±1.06 | 92.53 ±1.03 | 92.31 ±1.08 | 92.29 ±1.10 | 84.64 ±2.15 | 92.29 ±1.10 |
| GLCM-ELM (Pi and Lima 2021) | 74.19 ±2.74 | 77.81 ±2.03 | 77.01 ±1.29 | 76.00 ±0.98 | 75.54 ±1.31 | 52.08 ±1.95 | 75.57 ±1.28 |
| WE-Jaya (Wang 2021) | 73.31 ±2.26 | 78.11 ±1.92 | 77.03 ±1.35 | 75.71 ±1.04 | 75.10 ±1.23 | 51.51 ±2.07 | 75.14 ±1.22 |
| GLCM+SNN (Pi 2021) | 74.66 ±1.87 | 78.00 ±1.29 | 77.24 ±1.15 | 76.33 ±1.18 | 75.92 ±1.31 | 52.70 ±2.34 | 75.93 ±1.30 |
| WE-SAJ (Wang, Zhang et al. 2022) | 85.47 ±1.84 | 87.23 ±1.67 | 87.03 ±1.34 | 86.35 ±0.70 | 86.23 ±0.77 | 72.75 ±1.38 | 86.24 ±0.76 |
| PSTCNN (Ours) | 93.65 ±1.86 | 94.32 ±2.07 | 94.30 ±2.04 | 93.99 ±1.78 | 93.97 ±1.78 | 87.99 ±3.56 | 93.97 ±1.78 |
Sen: Sensitivity; Spc: Specificity; Prc: Precision; Acc: Accuracy; F1: F1-score; Mcc: Matthews correlation coefficient; FMI: Fowlkes-Mallows index.