Table 1.
Algorithms used in few state-of-the-art methods with its performance
| Schemes | Data size | Features | Representation | Classifier | Evaluation results |
|---|---|---|---|---|---|
| Vidal Plácido et al. (2020) | 179 | Texture and intensity | – | SVM-RBF | Accuracy: 83.83% |
| Srinivasan et al. (2014) | 42 | HOG | – | Linear-SVM | Accuracy: 86.7% |
| Liu et al. (2011) | 326 | Edge, LBP | PCA | SVM-RBF | AUC: 0.93 |
| Venhuizen et al. (2015) | 384 | Texton | BoW, PCA | Random forest | AUC:0.984 |
| Albarrak et al. (2013) | 140 | LBP-TOP, HoG | PCA | Bayesian network | Sensitivity: 87.4%, Specificity: 85.5% |
| Lemaître et al. (2015) | 32 | LBP, LBP-TOP | PCA, BoW, Histogram | Random forest | Sensitivity: 87.5%, Specificity: 75% |
| Sidibé et al. (2016) | 32 | Pixel-intensities | PCA | Mahalanobis distance to GMM | Sensitivity: 80%, Specificity: 93% |