EURASIP Journal on Image and Video Processing

Table 9 Classification results on the proposed feature selection approach for apple rust

From: A probabilistic segmentation and entropy-rank correlation-based feature selection approach for the recognition of fruit diseases

Method	FPR	Precision (%)	Sensitivity (%)	Specificity (%)	FNR (%)	Accuracy
DT	0.157	84.90	84.30	89.1	14.9	85.1
Q-SVM	0.096	90.05	90.30	90.9	9.6	93.6
C-SVM	0.070	92.15	92.90	90.9	7.4	92.6
QDA	0.157	83.95	84.15	70.9	18.1	81.9
F-KNN	0.061	93.2	93.8	92.7	6.4	93.6
W-KNN	0.103	88.50	89.60	81.8	11.7	88.3
EBT	0.096	90.05	90.30	90.9	9.6	92.6
ESDA	0.087	91.2	91.2	92.2	8.5	92.6
M-SVM	0.048	94.30	95.50	92.5	5.3	94.7

The bold values indicate best results

Back to article page