Table 9 Classification accuracy of models calibrated by individual pig herd within a scoring scale ≤ ±0.50 in each pig herd

From: Rapid construction method for a precision pork color scoring model based on standard color board images

Models

Accuracy (%)

Different pig herds

LW (N = 294)

LL (N = 89)

PM (N = 32)

Su (N = 47)

DLL (N = 50)

Overall (N = 512)

CS_1_ L*

95.92

93.26

93.75

80.85

94.00

93.75

CS_2_ L*

96.26

88.76

93.75

78.72

98.00

93.36

CS_3_ L*

93.20

93.26

93.75

78.72

94.00

91.99

CS_1_ L*a*

98.30

95.51

96.88

82.98

94.00

95.90

CS_2_ L*a*

99.32

97.75

100.00

91.49

98.00

98.24

CS_3_ L*a*

98.64

98.88

100.00

91.49

94.00

97.66

CS_1_ L*a*b*

95.24

100.00

96.88

93.62

96.00

96.10

CS_2_ L*a*b*

98.98

100.00

100.00

93.62

96.00

98.44

CS_3_ L*a*b*

98.98

100.00

100.00

93.62

98.00

98.63

  1. N It indicates the number of samples in each pig herd; The sample sizes of the LM and Me pig populations are small and do not have statistical significance, thus the pig herd for accuracies analysis of were excluded, LW Large White pigs, LL Landrace × Large White pigs, Su Suhuai pigs, PM Pietrain × Meishan pigs, DLL Duroc × (Landrace × Large White pigs).