Skip to main content

Table 10 Classification result on each extracted set of features for grapes rot leave disease

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

Method

Features

Performance measures

 

SLBP

Harlick

Color

Sen (%)

Spec (%)

Accuracy (%)

DT

  

89.35

91.0

89.4

  

 

86.40

86.6

86.4

   

93.95

92.5

94.1

QSVM

  

89.45

85.1

89.4

  

 

89.45

86.6

89.4

   

94.55

90.6

94.9

CSVM

  

90.10

91.0

90.2

  

 

87.90

86.6

87.9

   

94.55

90.6

94.9

QDA

  

90.30

80.6

90.2

  

 

88.0

80.6

87.9

   

93.95

92.5

94.1

Fine KNN

  

88.70

83.6

88.6

  

 

87.15

85.1

87.1

   

95.50

92.5

95.8

WKNN

  

91.75

86.6

91.7

  

 

82.60

82.4

82.6

   

94.70

92.5

94.9

EBT

  

90.85

86.6

90.9

  

 

89.45

86.6

89.4

   

95.30

90.6

95.8

ESDA

  

89.45

85.1

89.4

  

 

87.90

82.1

87.9

   

93.50

92.5

94.1

Multi-class SVM

  

93.15

92.5

93.2

  

 

91.0

85.1

90.9

   

95.30

90.6

95.8

  1. The bold values indicate best results