Skip to main content

Table 2 The average performance of baseline multi-class CNN models with different evaluation metrics on images without soil background

From: Deep learning-based detection of seedling development

Model

Accuracy

Error

Sensitivity

Specificity

Precision

False positive rate

Proposed CNN

0.80 ± 0.19

0.20 ± 0.19

0.85 ± 0.13

0.93 ± 0.07

0.85 ± 0.14

0.07 ± 0.07

VGG16

0.80 ± 0.24

0.2 ± 0.24

0.84 ± 0.18

0.93 ± 0.12

0.85 ± 0.07

0.07 ± 0.11

ResNet50

0.78 ± 0.18

0.22 ± 0.18

0.77 ± 0.21

0.89 ± 0.09

0.85 ± 0.11

0.08 ± 0.05

DenseNet121

0.79 ± 0.09

0.21 ±  0.09

0.78 ±  0.08

0.90 ±  0.14

0.86 ±  0.09

0.07 ±  0.10