Skip to main content

Table 10 LDA Plankton images

From: Deriving image features for autonomous classification from time-series recurrence plots

STANDARD Annelida Appendicularia Bacillariophyceae Cnidaria Crustacea Dinoflagellata Marine snow Mollusca Vertebrata Prediction success
Annelida 15 1 0 2 30 0 0 0 2 0.30
Appendicularia 0 37 0 0 8 1 0 0 4 0.74
Bacillariophyceae 0 0 34 0 8 16 37 5 0 0.34
Cnidaria 8 9 0 29 8 2 3 1 0 0.48
Crustacea 21 2 6 4 455 0 15 4 3 0.89
Dinoflagellata 0 1 3 0 0 38 0 3 0 0.84
Marine snow 4 2 23 0 103 0 18 0 0 0.12
Mollusca 0 0 8 0 6 3 2 9 0 0.32
Vertebrata 3 10 0 0 8 0 0 0 4 0.16
Model           0.628
RQA           
Annelida 5 1 0 15 29 0 0 0 0 0.10
Appendicularia 0 24 0 0 23 0 0 0 3 0.48
Bacillariophyceae 0 0 35 0 22 26 12 5 0 0.35
Cnidaria 6 2 0 13 33 0 0 0 6 0.22
Crustacea 7 6 10 16 432 0 28 0 11 0.85
Dinoflagellata 0 0 12 0 6 22 4 1 0 0.49
Marine snow 0 0 14 3 107 8 17 1 0 0.11
Mollusca 0 0 4 0 19 2 3 0 0 0.00
Vertebrata 0 8 0 1 4 0 0 0 12 0.48
Model           0.550
STANDARD & RQA           
Annelida 22 1 0 3 24 0 0 0 0 0.44
Appendicularia 0 39 0 0 6 0 0 0 5 0.78
Bacillariophyceae 1 0 48 0 3 20 20 8 0 0.48
Cnidaria 10 1 1 39 5 0 1 2 1 0.65
Crustacea 18 3 6 2 437 0 33 7 4 0.86
Dinoflagellata 0 1 3 0 0 34 0 7 0 0.76
Marine snow 2 2 20 1 86 1 33 5 0 0.22
Mollusca 0 0 5 0 6 2 4 11 0 0.39
Vertebrata 1 7 0 0 7 0 0 0 10 0.40
Model           0.661
STANDARD & RQA & ESQ           
Annelida 24 1 0 3 22 0 0 0 0 0.48
Appendicularia 0 40 0 0 5 0 0 0 5 0.80
Bacillariophyceae 1 0 47 0 2 21 22 7 0 0.47
Cnidaria 6 1 0 36 6 0 4 4 3 0.60
Crustacea 19 4 5 2 436 0 30 7 7 0.85
Dinoflagellata 0 1 3 0 0 33 0 8 0 0.73
Marine snow 2 2 19 1 89 1 31 5 0 0.20
Mollusca 0 0 4 0 5 2 5 12 0 0.43
Vertebrata 1 7 0 0 6 0 0 0 11 0.44
Model           0.658
  1. Confusion matrices of the different parameter combinations used for the LDA analyses
  2. Values in bold represent true positive classifications of the LDA model and overall discriminative success