Table 5.
Comparison of accuracy (%) of PGLCM-SARF and PGLCM-RF.
| Category | Model | Magnification | ||||
|---|---|---|---|---|---|---|
| Irrelevant | 40X | 100X | 200X | 400X | ||
| Eight-class Classification | Proposed method | 92.67 | 92.79 | 93.72 | 94.16 | 93.67 |
| PGLCM-RF | 92.63 | 92.63 | 93.02 | 93.96 | 93.18 | |
| Difference | 0.04 | 0.16 | 0.70 | 0.20 | 0.49 | |
| Binary Classification | Proposed method | 97.11 | 97.56 | 97.69 | 97.76 | 97.54 |
| PGLCM-RF | 96.54 | 96.94 | 97.01 | 97.21 | 97.17 | |
| Difference | 0.22 | 0.62 | 0.68 | 0.55 | 0.37 | |
The bold values indicate the highest accuracy under the same conditions.