Skip to main content

Table 8 The classification results (%) of the TL method on our LDWBC test set

From: Accurate classification of white blood cells by coupling pre-trained ResNet and DenseNet with SCAM mechanism

Our model

OA

AP

AR

AF1

RM

DM

    

TL

No TL

TL

No TL

    

\(\checkmark\)

  

\(\checkmark\)

97.84

91.30

95.35

93.21

 

\(\checkmark\)

\(\checkmark\)

 

97.46

92.30

93.04

92.45

 

\(\checkmark\)

 

\(\checkmark\)

96.20

92.34

82.64

85.88

\(\checkmark\)

 

\(\checkmark\)

 

97.84

91.61

96.38

93.82

  1. Best results are in bold; RM ResNet module, DM DenseNet module