Skip to main content
. 2018 Jun 18;8:9317. doi: 10.1038/s41598-018-27586-9

Table 2.

Classification accuracy of automatic classification system.

Feature Method Performance on the testing set
Sensitivity Specificity
Global geometrical features PCA + SVM 4/16 = 25% 39/48 = 81.3%
Global texture features PCA + SVM 6/16 = 37.5% 42/48 = 87.5%
Fusion of local features AdaBoost 11/16 = 68.8% 42/48 = 87.5%

SVM, support vector machine.

PCA, principal component analysis.