Page 184 - Handbook of Deep Learning in Biomedical Engineering Techniques and Applications
P. 184
Chapter 6 Plant leaf disease classification based on feature selection 173
(A) Training vs Valida on Loss
250
200
150
Loss
100
50
0
1 2 3 4 5 6 7 8 9 10111213141516171819202122232425
Epoch
Training loss Valida on loss
(B) Training vs Valida on Accuracy
80%
70%
60%
50%
Accuracy 40%
30%
20%
10%
0%
1 2 3 4 5 6 7 8 9 10111213141516171819202122232425
Epoch
Training acc Valida on acc
Figure 6.15 Training and validation result of ResNet-50 model.
The features considered in this study include the following:
• Statistical features: mean, standard deviation, skewness, and
kurtosis of color channels (R, G, B, H, S, V).
• Geometric features:
• Defect area
• Defect perimeter: measured by the length of defected
region boundary
