Table 12 Confusion matrix for all set of extracted features using ISIC-UDA dataset
Class | Total images | Melanoma | Benign |
|---|---|---|---|
Confusion matrix: Proposed features fusion and selection | |||
Melanoma | 93 | 95.7% | 4.3% |
Benign | 140 | - | 100% |
Confusion matrix: Harlick features | |||
Melanoma | 93 | 55.9% | 44.1% |
Benign | 140 | 8.6% | 91.4% |
Confusion matrix: HOG features | |||
Melanoma | 93 | 68.8% | 31.2% |
Benign | 140 | 3.6% | 96.4% |
Confusion matrix: Color features | |||
Melanoma | 93 | 92.5% | 7.5% |
Benign | 140 | 2.1% | 97.9% |