Blended Features Classification of Leaf-Based Cucumber Disease Using Image Processing Techniques

Table 5

Performance evaluation of experiment 2.

Classifier	Performance evaluation
	Accuracy (%)	Specificity (%)	Sensitivity (%)	Precision (%)	FNR (%)	FPR
Fine KNN	94.6	84.0	84.0	84.6	5.4	0.011
Subspace KNN	94.5	83.8	83.8	84.3	5.6	0.011
Bagged trees	94.2	80.5	80.5	81.0	5.8	0.013
Weighted KNN	93.6	81.0	81.0	82.5	6.4	0.013
SVM cubic	91.6	90.8	90.8	93.6	8.4	0.018
Fine tree	81.8	76.6	76.6	81.6	18.2	0.045

Bold shows the highest values in terms of accuracy. The Fine KNN classifier has the highest accuracy of 94.6%, specificity of 84.0%, sensitivity of 84.0%, precision of 84.6%, FNR rate of 5.4, and FPR is 0.011.