Table 3 Overview of the ablation studies.

From: A dual-track feature fusion model utilizing Group Shuffle Residual DeformNet and swin transformer for the classification of grape leaf diseases

Model architecture

Accuracy (in %)

Precision (in %)

Recall (in %)

F1 Score (in %)

Swin transformer

95.27

95.37

95.64

95.48

Proposed CNN without novel blocks

93.12

93.2

93.48

93.24

Proposed CNN without DC

94.41

94.82

94.48

94.51

Group Shuffle Residual DeformNet

96.02

96.5

95.85

96.13

Proposed network without triplet attention

98.17

98.32

98.21

98.27

Proposed network

98.6

98.7

98.59

98.64

Back to article page