Skip to main content

Table 3 Confusion matrix of the percentages of pixel classification for each leaf class, i.e. stresses sourced by Fusarium oxysporum f.sp. raphani (F), Rhizoctonia solani (R), salinity (S) and water deficit (W) compared to control (C) treatment

From: Sorting biotic and abiotic stresses on wild rocket by leaf-image hyperspectral data mining with an artificial intelligence model

Treatments

Output

C

F

R

S

W

C

0.34 ± 0.10a

0.14 ± 0.07bc

0.12 ± 0.06c

0.19 ± 0.09b

0.21 ± 0.10b

F

0.05 ± 0.06c

0.56 ± 0.18a

0.16 ± 0.10b

0.06 ± 0.08c

0.18 ± 0.09b

R

0.08 ± 0.11c

0.22 ± 0.13b

0.45 ± 0.19a

0.08 ± 0.09c

0.18 ± 0.11b

S

0.16 ± 0.11bc

0.15 ± 0.08c

0.14 ± 0.08c

0.34 ± 0.08a

0.20 ± 0.09b

W

0.10 ± 0.07c

0.21 ± 0.14b

0.20 ± 0.13b

0.12 ± 0.09c

0.37 ± 0.12a

  1. ANOVA test (H0: same mean) results are reported using letters. Horizontally reading, equal letters correspond to no significant differences
  2. Observing the variable impact of spectral values, the most informative ones ranged within the following frequencies: 492–504 nm, 540–568 nm, 712–720 nm, 855 nm, 900–908 nm and 970 nm (Fig. 6)
Back to article page

Plant Methods

ISSN: 1746-4811