Table 4 Test for accuracy of the DL (deep learning) process for the Culex (Linnaeus, 1758) species identification, in percentage of accuracy.

From: Application of wings interferential patterns (WIPs) and deep learning (DL) to classify some Culex. spp (Culicidae) of medical or veterinary importance

     

Predicted

     
 

Species

ins

neav

neb

quinq

thal

trit

univ

Others

Nb

 

ins

75.0

0.0

0.0

0.0

0.0

0.0

0.0

25.0

4

T

neav

0.0

100.0

0.0

0.0

0.0

0.0

0.0

0.0

52

r

neb

0.0

0.0

50.0

0.0

0.0

0.0

0.0

50.0

2

u

quinq

0.0

7.1

0.0

92.9

0.0

0.0

0.0

0.0

14

t

thal

0.0

0.0

0.0

0.0

0.0

0.0

0.0

100.0

2

h

trit

0.0

0.0

0.0

2.9

0.0

94.1

0.0

2.9

34

 

Univ

0.0

40.0

0.0

0.0

0.0

20.0

40.0

0.0

5

 

Others

0.0

0.6

0.1

0.1

0.0

0.1

0.0

99.1

905

  1. Cx. insignis; neav, Cx. neavei; neb, Cx. nebulosus; quinq, Cx. quinquefaciatus; thal, Cx. thalassius; trit, Cx tritaenorhynchus; univ, Cx. univittatus; Others, other culicidae species not belonging to the Culex genus; Nb, number of pictures tested.
Back to article page