Figures (5)  Tables (2)
    • Figure 1. 

      Establishment of PLS-DA model for accurate identification of Chinese cabbage leaf color based on multispectral imaging.

    • Figure 2. 

      Visual assessment of leaf color phenotype. (a) The leaf color phenotypes of nine homozygous high-generation inbred lines and their F1 generation. The figure references the Grass green, Emerald green, May green, and Yellow green in 'The Royal Horticultural Society's Color Chart' as color comparisons. Representative materials of (b) 'F2-442' and (c) 'F2-449' in F2 population of Chinese cabbage. (d) Visual inspection on the difference of leaf color of population. (e) Color difference classification of samples. Scale bars = 10 cm.

    • Figure 3. 

      Comparison of leaf color identification accuracy of Chinese cabbage based on K-means. Spatial distribution of (a) RGB parameters and (b) HSL parameter identification results. (c) K-means clustering recognition rate of RGB parameters, HSL parameters and multispectral data.

    • Figure 4. 

      PLS-DA model parameters. (a) Model fitting index for each sample type on image RGB, HSL color parameters, multispectral data and data fusion. (b) Overall recognition rate of RGB parameters, HSL parameters, multispectral data and data fusion.

    • Figure 5. 

      Display the distribution map of SNP index on chromosomes using Gprime's computational model. The Gprime analysis results of BSA sequencing for identifying 'F2-449' population leaf color classification results based on (a) visual inspection and (b) PLS-DA fusion model (b).

    • Principal
      divisor
      RGBHSLMultispectralRGB + HSL+ Multispectral
      123Count12Count1234Count1234567Count
      R2X0.8970.0940.00810.7510.2420.9930.6590.2340.080.0090.9820.6910.1910.0650.0350.0050.0040.0040.995
      R2Y0.2360.0280.0740.3390.2360.1080.3440.2360.1960.0240.0270.4830.2380.1980.0420.0250.5370.0790.0470.663
      Q20.2310.0050.0830.2990.2330.1330.3350.2310.2480.0090.0030.4290.2330.2520.0420.0230.050.0620.0870.546
      Root-mean-
      square error
      (RMSE)
      A0.2160.2120.210.2130.2410.1910.2160.2330.2030.1990.2020.2090.2280.1990.1990.20.2160.2150.20.208
      B0.3760.3790.3530.3690.3910.3350.3630.3860.3070.3120.310.3290.3850.3050.3070.3040.310.3180.3230.322
      C0.4080.4140.410.4110.4050.3980.4020.4030.3520.3590.3610.3690.4050.3540.360.3610.3760.3840.3810.374
      D0.3330.3290.2760.3130.2980.2720.2850.320.2860.270.260.2840.1950.2240.2450.2360.2450.2770.320.249
      E0.390.3810.3850.3850.3890.390.390.3870.3510.2480.3470.3330.3880.3550.3430.3340.3480.3330.3070.344

      Table 1. 

      Accuracy of PLS-DA model based on image RGB color parameters, HSL color parameters, multispectral data, and data fusion.

    • Leaf color recognition methodChromosomePositionSNV number
      Manual visual inspectionChr 949732120-604949643450
      Data fusion-based PLS-DA leaf color identificationChr 953002792-539998991539

      Table 2. 

      Candidate region statistics.